CINXE.COM

<!DOCTYPE html> <html > <head> <meta charset="utf-8"> <meta rel="search" type="application/opensearchdescription+xml" href="/open_search.xml" title="Academia.edu"> <meta content="width=device-width, initial-scale=1" name="viewport"> <meta name="google-site-verification" content="bKJMBZA7E43xhDOopFZkssMMkBRjvYERV-NaN4R6mrs"> <meta name="csrf-param" content="authenticity_token" /> <meta name="csrf-token" content="eEBsNCWocPL/rRK+qjGoJhJUNT39WgGDYiTs+/mBHv42G9q1Z9wj0rlXB0W7bwf0X1Q6fT6jZysdU6NSwTjPTA==" /> <meta name="citation_title" content="Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words" /> <meta name="citation_journal_title" content="ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)" /> <meta name="citation_author" content="Santiago Cuervo" /> <meta name="twitter:card" content="summary" /> <meta name="twitter:url" content="https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words" /> <meta name="twitter:title" content="Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words" /> <meta name="twitter:description" content="Academia.edu is a platform for academics to share research papers." /> <meta name="twitter:image" content="http://a.academia-assets.com/images/twitter-card.jpeg" /> <meta property="fb:app_id" content="2369844204" /> <meta property="og:type" content="article" /> <meta property="og:url" content="https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words" /> <meta property="og:title" content="Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words" /> <meta property="og:image" content="http://a.academia-assets.com/images/open-graph-icons/fb-paper.gif" /> <meta property="og:description" content="Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words" /> <meta property="article:author" content="https://independent.academia.edu/SantiagoCuervo19" /> <meta name="description" content="Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words" /> <title>(PDF) Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words | Santiago Cuervo - Academia.edu</title> <link rel="canonical" href="https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words" /> <script async src="https://www.googletagmanager.com/gtag/js?id=G-5VKX33P2DS"></script> <script> window.dataLayer = window.dataLayer || []; function gtag(){dataLayer.push(arguments);} gtag('js', new Date()); gtag('config', 'G-5VKX33P2DS', { cookie_domain: 'academia.edu', send_page_view: false, }); gtag('event', 'page_view', { 'controller': "single_work", 'action': "show", 'controller_action': 'single_work#show', 'logged_in': 'false', 'edge': 'unknown', // Send nil if there is no A/B test bucket, in case some records get logged // with missing data - that way we can distinguish between the two cases. // ab_test_bucket should be of the form <ab_test_name>:<bucket> 'ab_test_bucket': null, }) </script> <script> var $controller_name = 'single_work'; var $action_name = "show"; var $rails_env = 'production'; var $app_rev = 'b55351076b26328bd3ebb8c0edd1b35967bcfaed'; var $domain = 'academia.edu'; var $app_host = "academia.edu"; var $asset_host = "academia-assets.com"; var $start_time = new Date().getTime(); var $recaptcha_key = "6LdxlRMTAAAAADnu_zyLhLg0YF9uACwz78shpjJB"; var $recaptcha_invisible_key = "6Lf3KHUUAAAAACggoMpmGJdQDtiyrjVlvGJ6BbAj"; var $disableClientRecordHit = false; </script> <script> window.require = { config: function() { return function() {} } } </script> <script> window.Aedu = window.Aedu || {}; window.Aedu.hit_data = null; window.Aedu.serverRenderTime = new Date(1732751279000); window.Aedu.timeDifference = new Date().getTime() - 1732751279000; </script> <script type="application/ld+json">{"@context":"https://schema.org","@type":"ScholarlyArticle","abstract":null,"author":[{"@context":"https://schema.org","@type":"Person","name":"Santiago Cuervo"}],"contributor":[],"dateCreated":"2022-11-15","dateModified":"2024-11-24","datePublished":null,"headline":"Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words","inLanguage":"en","keywords":["Engineering","Computer Science","Artificial Intelligence","Speech Recognition","Categorization","Segmentation","arXiv"],"locationCreated":null,"publication":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","publisher":{"@context":"https://schema.org","@type":"Organization","name":"IEEE"},"image":null,"thumbnailUrl":null,"url":"https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words","sourceOrganization":[{"@context":"https://schema.org","@type":"EducationalOrganization","name":null}]}</script><link rel="stylesheet" media="all" href="//a.academia-assets.com/assets/single_work_page/loswp-102fa537001ba4d8dcd921ad9bd56c474abc201906ea4843e7e7efe9dfbf561d.css" /><link rel="stylesheet" media="all" href="//a.academia-assets.com/assets/design_system/body-8d679e925718b5e8e4b18e9a4fab37f7eaa99e43386459376559080ac8f2856a.css" /><link rel="stylesheet" media="all" href="//a.academia-assets.com/assets/design_system/button-3cea6e0ad4715ed965c49bfb15dedfc632787b32ff6d8c3a474182b231146ab7.css" /><link rel="stylesheet" media="all" href="//a.academia-assets.com/assets/design_system/text_button-73590134e40cdb49f9abdc8e796cc00dc362693f3f0f6137d6cf9bb78c318ce7.css" /><link crossorigin="" href="https://fonts.gstatic.com/" rel="preconnect" /><link href="https://fonts.googleapis.com/css2?family=DM+Sans:ital,opsz,wght@0,9..40,100..1000;1,9..40,100..1000&family=Gupter:wght@400;500;700&family=IBM+Plex+Mono:wght@300;400&family=Material+Symbols+Outlined:opsz,wght,FILL,GRAD@20,400,0,0&display=swap" rel="stylesheet" /><link rel="stylesheet" media="all" href="//a.academia-assets.com/assets/design_system/common-10fa40af19d25203774df2d4a03b9b5771b45109c2304968038e88a81d1215c5.css" /> </head> <body> <div id='react-modal'></div> <div class="js-upgrade-ie-banner" style="display: none; text-align: center; padding: 8px 0; background-color: #ebe480;"><p style="color: #000; font-size: 12px; margin: 0 0 4px;">Academia.edu no longer supports Internet Explorer.</p><p style="color: #000; font-size: 12px; margin: 0;">To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to <a href="https://www.academia.edu/upgrade-browser">upgrade your browser</a>.</p></div><script>// Show this banner for all versions of IE if (!!window.MSInputMethodContext || /(MSIE)/.test(navigator.userAgent)) { document.querySelector('.js-upgrade-ie-banner').style.display = 'block'; }</script> <div class="bootstrap login"><div class="modal fade login-modal" id="login-modal"><div class="login-modal-dialog modal-dialog"><div class="modal-content"><div class="modal-header"><button class="close close" data-dismiss="modal" type="button"><span aria-hidden="true">×</span><span class="sr-only">Close</span></button><h4 class="modal-title text-center"><strong>Log In</strong></h4></div><div class="modal-body"><div class="row"><div class="col-xs-10 col-xs-offset-1"><button class="btn btn-fb btn-lg btn-block btn-v-center-content" id="login-facebook-oauth-button"><svg style="float: left; width: 19px; line-height: 1em; margin-right: .3em;" aria-hidden="true" focusable="false" data-prefix="fab" data-icon="facebook-square" class="svg-inline--fa fa-facebook-square fa-w-14" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512"><path fill="currentColor" d="M400 32H48A48 48 0 0 0 0 80v352a48 48 0 0 0 48 48h137.25V327.69h-63V256h63v-54.64c0-62.15 37-96.48 93.67-96.48 27.14 0 55.52 4.84 55.52 4.84v61h-31.27c-30.81 0-40.42 19.12-40.42 38.73V256h68.78l-11 71.69h-57.78V480H400a48 48 0 0 0 48-48V80a48 48 0 0 0-48-48z"></path></svg><small><strong>Log in</strong> with <strong>Facebook</strong></small></button><br /><button class="btn btn-google btn-lg btn-block btn-v-center-content" id="login-google-oauth-button"><svg style="float: left; width: 22px; line-height: 1em; margin-right: .3em;" aria-hidden="true" focusable="false" data-prefix="fab" data-icon="google-plus" class="svg-inline--fa fa-google-plus fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M256,8C119.1,8,8,119.1,8,256S119.1,504,256,504,504,392.9,504,256,392.9,8,256,8ZM185.3,380a124,124,0,0,1,0-248c31.3,0,60.1,11,83,32.3l-33.6,32.6c-13.2-12.9-31.3-19.1-49.4-19.1-42.9,0-77.2,35.5-77.2,78.1S142.3,334,185.3,334c32.6,0,64.9-19.1,70.1-53.3H185.3V238.1H302.2a109.2,109.2,0,0,1,1.9,20.7c0,70.8-47.5,121.2-118.8,121.2ZM415.5,273.8v35.5H380V273.8H344.5V238.3H380V202.8h35.5v35.5h35.2v35.5Z"></path></svg><small><strong>Log in</strong> with <strong>Google</strong></small></button><br /><style type="text/css">.sign-in-with-apple-button { width: 100%; height: 52px; border-radius: 3px; border: 1px solid black; cursor: pointer; }</style><script src="https://appleid.cdn-apple.com/appleauth/static/jsapi/appleid/1/en_US/appleid.auth.js" type="text/javascript"></script><div class="sign-in-with-apple-button" data-border="false" data-color="white" id="appleid-signin"><span   ="Sign Up with Apple" class="u-fs11"></span></div><script>AppleID.auth.init({ clientId: 'edu.academia.applesignon', scope: 'name email', redirectURI: 'https://www.academia.edu/sessions', state: "658d8d0c0756fe8d117b21051a82052c1331ab73f6f7552e5dfde9aabc9456a2", });</script><script>// Hacky way of checking if on fast loswp if (window.loswp == null) { (function() { const Google = window?.Aedu?.Auth?.OauthButton?.Login?.Google; const Facebook = window?.Aedu?.Auth?.OauthButton?.Login?.Facebook; if (Google) { new Google({ el: '#login-google-oauth-button', rememberMeCheckboxId: 'remember_me', track: null }); } if (Facebook) { new Facebook({ el: '#login-facebook-oauth-button', rememberMeCheckboxId: 'remember_me', track: null }); } })(); }</script></div></div></div><div class="modal-body"><div class="row"><div class="col-xs-10 col-xs-offset-1"><div class="hr-heading login-hr-heading"><span class="hr-heading-text">or</span></div></div></div></div><div class="modal-body"><div class="row"><div class="col-xs-10 col-xs-offset-1"><form class="js-login-form" action="https://www.academia.edu/sessions" accept-charset="UTF-8" method="post"><input name="utf8" type="hidden" value="✓" autocomplete="off" /><input type="hidden" name="authenticity_token" value="OyhdEJtzn91S8TG0OL88Ioz7IEWKK9pPxDlcVkQm8391c+uR2QfM/RQLJE8p4ZPwwfsvBUnSvOe7ThP/fJ8izQ==" autocomplete="off" /><div class="form-group"><label class="control-label" for="login-modal-email-input" style="font-size: 14px;">Email</label><input class="form-control" id="login-modal-email-input" name="login" type="email" /></div><div class="form-group"><label class="control-label" for="login-modal-password-input" style="font-size: 14px;">Password</label><input class="form-control" id="login-modal-password-input" name="password" type="password" /></div><input type="hidden" name="post_login_redirect_url" id="post_login_redirect_url" value="https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words" autocomplete="off" /><div class="checkbox"><label><input type="checkbox" name="remember_me" id="remember_me" value="1" checked="checked" /><small style="font-size: 12px; margin-top: 2px; display: inline-block;">Remember me on this computer</small></label></div><br><input type="submit" name="commit" value="Log In" class="btn btn-primary btn-block btn-lg js-login-submit" data-disable-with="Log In" /></br></form><script>typeof window?.Aedu?.recaptchaManagedForm === 'function' && window.Aedu.recaptchaManagedForm( document.querySelector('.js-login-form'), document.querySelector('.js-login-submit') );</script><small style="font-size: 12px;"><br />or <a data-target="#login-modal-reset-password-container" data-toggle="collapse" href="javascript:void(0)">reset password</a></small><div class="collapse" id="login-modal-reset-password-container"><br /><div class="well margin-0x"><form class="js-password-reset-form" action="https://www.academia.edu/reset_password" accept-charset="UTF-8" method="post"><input name="utf8" type="hidden" value="✓" autocomplete="off" /><input type="hidden" name="authenticity_token" value="1MTEQ7ZNIGcMeroS/i5Usvk0qfQwhD0KpxFewBIfDrOan3LC9DlzR0qAr+nvcPtgtDSmtPN9W6LYZhFpKqbfAQ==" autocomplete="off" /><p>Enter the email address you signed up with and we'll email you a reset link.</p><div class="form-group"><input class="form-control" name="email" type="email" /></div><input class="btn btn-primary btn-block g-recaptcha js-password-reset-submit" data-sitekey="6Lf3KHUUAAAAACggoMpmGJdQDtiyrjVlvGJ6BbAj" type="submit" value="Email me a link" /></form></div></div><script> require.config({ waitSeconds: 90 })(["https://a.academia-assets.com/assets/collapse-45805421cf446ca5adf7aaa1935b08a3a8d1d9a6cc5d91a62a2a3a00b20b3e6a.js"], function() { // from javascript_helper.rb $("#login-modal-reset-password-container").on("shown.bs.collapse", function() { $(this).find("input[type=email]").focus(); }); }); </script> </div></div></div><div class="modal-footer"><div class="text-center"><small style="font-size: 12px;">Need an account? <a rel="nofollow" href="https://www.academia.edu/signup">Click here to sign up</a></small></div></div></div></div></div></div><script>// If we are on subdomain or non-bootstrapped page, redirect to login page instead of showing modal (function(){ if (typeof $ === 'undefined') return; var host = window.location.hostname; if ((host === $domain || host === "www."+$domain) && (typeof $().modal === 'function')) { $("#nav_log_in").click(function(e) { // Don't follow the link and open the modal e.preventDefault(); $("#login-modal").on('shown.bs.modal', function() { $(this).find("#login-modal-email-input").focus() }).modal('show'); }); } })()</script> <div id="fb-root"></div><script>window.fbAsyncInit = function() { FB.init({ appId: "2369844204", version: "v8.0", status: true, cookie: true, xfbml: true }); // Additional initialization code. if (window.InitFacebook) { // facebook.ts already loaded, set it up. window.InitFacebook(); } else { // Set a flag for facebook.ts to find when it loads. window.academiaAuthReadyFacebook = true; } };</script> <div id="google-root"></div><script>window.loadGoogle = function() { if (window.InitGoogle) { // google.ts already loaded, set it up. window.InitGoogle("331998490334-rsn3chp12mbkiqhl6e7lu2q0mlbu0f1b"); } else { // Set a flag for google.ts to use when it loads. window.GoogleClientID = "331998490334-rsn3chp12mbkiqhl6e7lu2q0mlbu0f1b"; } };</script> <div class="header--container" id="main-header-container"><div class="header--inner-container header--inner-container-ds2"><div class="header-ds2--left-wrapper"><div class="header-ds2--left-wrapper-inner"><a data-main-header-link-target="logo_home" href="https://www.academia.edu/"><img class="hide-on-desktop-redesign" style="height: 24px; width: 24px;" alt="Academia.edu" src="//a.academia-assets.com/images/academia-logo-redesign-2015-A.svg" width="24" height="24" /><img width="145.2" height="18" class="hide-on-mobile-redesign" style="height: 24px;" alt="Academia.edu" src="//a.academia-assets.com/images/academia-logo-redesign-2015.svg" /></a><div class="header--search-container header--search-container-ds2"><form class="js-SiteSearch-form select2-no-default-pills" action="https://www.academia.edu/search" accept-charset="UTF-8" method="get"><input name="utf8" type="hidden" value="✓" autocomplete="off" /><svg style="width: 14px; height: 14px;" aria-hidden="true" focusable="false" data-prefix="fas" data-icon="search" class="header--search-icon svg-inline--fa fa-search fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M505 442.7L405.3 343c-4.5-4.5-10.6-7-17-7H372c27.6-35.3 44-79.7 44-128C416 93.1 322.9 0 208 0S0 93.1 0 208s93.1 208 208 208c48.3 0 92.7-16.4 128-44v16.3c0 6.4 2.5 12.5 7 17l99.7 99.7c9.4 9.4 24.6 9.4 33.9 0l28.3-28.3c9.4-9.4 9.4-24.6.1-34zM208 336c-70.7 0-128-57.2-128-128 0-70.7 57.2-128 128-128 70.7 0 128 57.2 128 128 0 70.7-57.2 128-128 128z"></path></svg><input class="header--search-input header--search-input-ds2 js-SiteSearch-form-input" data-main-header-click-target="search_input" name="q" placeholder="Search" type="text" /></form></div></div></div><nav class="header--nav-buttons header--nav-buttons-ds2 js-main-nav"><a class="ds2-5-button ds2-5-button--secondary js-header-login-url header-button-ds2 header-login-ds2 hide-on-mobile-redesign" href="https://www.academia.edu/login" rel="nofollow">Log In</a><a class="ds2-5-button ds2-5-button--secondary header-button-ds2 hide-on-mobile-redesign" href="https://www.academia.edu/signup" rel="nofollow">Sign Up</a><button class="header--hamburger-button header--hamburger-button-ds2 hide-on-desktop-redesign js-header-hamburger-button"><div class="icon-bar"></div><div class="icon-bar" style="margin-top: 4px;"></div><div class="icon-bar" style="margin-top: 4px;"></div></button></nav></div><ul class="header--dropdown-container js-header-dropdown"><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/login" rel="nofollow">Log In</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/signup" rel="nofollow">Sign Up</a></li><li class="header--dropdown-row js-header-dropdown-expand-button"><button class="header--dropdown-button">more<svg aria-hidden="true" focusable="false" data-prefix="fas" data-icon="caret-down" class="header--dropdown-button-icon svg-inline--fa fa-caret-down fa-w-10" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 320 512"><path fill="currentColor" d="M31.3 192h257.3c17.8 0 26.7 21.5 14.1 34.1L174.1 354.8c-7.8 7.8-20.5 7.8-28.3 0L17.2 226.1C4.6 213.5 13.5 192 31.3 192z"></path></svg></button></li><li><ul class="header--expanded-dropdown-container"><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/about">About</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/press">Press</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://medium.com/@academia">Blog</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/documents">Papers</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/terms">Terms</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/privacy">Privacy</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/copyright">Copyright</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://www.academia.edu/hiring"><svg aria-hidden="true" focusable="false" data-prefix="fas" data-icon="briefcase" class="header--dropdown-row-icon svg-inline--fa fa-briefcase fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M320 336c0 8.84-7.16 16-16 16h-96c-8.84 0-16-7.16-16-16v-48H0v144c0 25.6 22.4 48 48 48h416c25.6 0 48-22.4 48-48V288H320v48zm144-208h-80V80c0-25.6-22.4-48-48-48H176c-25.6 0-48 22.4-48 48v48H48c-25.6 0-48 22.4-48 48v80h512v-80c0-25.6-22.4-48-48-48zm-144 0H192V96h128v32z"></path></svg>We're Hiring!</a></li><li class="header--dropdown-row"><a class="header--dropdown-link" href="https://support.academia.edu/"><svg aria-hidden="true" focusable="false" data-prefix="fas" data-icon="question-circle" class="header--dropdown-row-icon svg-inline--fa fa-question-circle fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M504 256c0 136.997-111.043 248-248 248S8 392.997 8 256C8 119.083 119.043 8 256 8s248 111.083 248 248zM262.655 90c-54.497 0-89.255 22.957-116.549 63.758-3.536 5.286-2.353 12.415 2.715 16.258l34.699 26.31c5.205 3.947 12.621 3.008 16.665-2.122 17.864-22.658 30.113-35.797 57.303-35.797 20.429 0 45.698 13.148 45.698 32.958 0 14.976-12.363 22.667-32.534 33.976C247.128 238.528 216 254.941 216 296v4c0 6.627 5.373 12 12 12h56c6.627 0 12-5.373 12-12v-1.333c0-28.462 83.186-29.647 83.186-106.667 0-58.002-60.165-102-116.531-102zM256 338c-25.365 0-46 20.635-46 46 0 25.364 20.635 46 46 46s46-20.636 46-46c0-25.365-20.635-46-46-46z"></path></svg>Help Center</a></li><li class="header--dropdown-row js-header-dropdown-collapse-button"><button class="header--dropdown-button">less<svg aria-hidden="true" focusable="false" data-prefix="fas" data-icon="caret-up" class="header--dropdown-button-icon svg-inline--fa fa-caret-up fa-w-10" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 320 512"><path fill="currentColor" d="M288.662 352H31.338c-17.818 0-26.741-21.543-14.142-34.142l128.662-128.662c7.81-7.81 20.474-7.81 28.284 0l128.662 128.662c12.6 12.599 3.676 34.142-14.142 34.142z"></path></svg></button></li></ul></li></ul></div> <script src="//a.academia-assets.com/assets/webpack_bundles/fast_loswp-bundle-71e03f93a0fba43adc4297a781256a72e56b0d578ac299a0d81b09f4c7bc6f70.js" defer="defer"></script><script>window.loswp = {}; window.loswp.author = 140931819; window.loswp.bulkDownloadFilterCounts = {}; window.loswp.hasDownloadableAttachment = true; window.loswp.hasViewableAttachments = true; // TODO: just use routes for this window.loswp.loginUrl = "https://www.academia.edu/login?post_login_redirect_url=https%3A%2F%2Fwww.academia.edu%2F90855738%2FContrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words%3Fauto%3Ddownload"; window.loswp.translateUrl = "https://www.academia.edu/login?post_login_redirect_url=https%3A%2F%2Fwww.academia.edu%2F90855738%2FContrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words%3Fshow_translation%3Dtrue"; window.loswp.previewableAttachments = [{"id":94302544,"identifier":"Attachment_94302544","shouldShowBulkDownload":false}]; window.loswp.shouldDetectTimezone = true; window.loswp.shouldShowBulkDownload = true; window.loswp.showSignupCaptcha = false window.loswp.willEdgeCache = false; window.loswp.work = {"work":{"id":90855738,"created_at":"2022-11-15T11:52:58.045-08:00","from_world_paper_id":220371400,"updated_at":"2024-11-24T15:40:45.393-08:00","_data":{"publisher":"IEEE","grobid_abstract":"We investigate the performance on phoneme categorization and phoneme and word segmentation of several selfsupervised learning (SSL) methods based on Contrastive Predictive Coding (CPC). Our experiments show that with the existing algorithms there is a trade off between categorization and segmentation performance. We investigate the source of this conflict and conclude that the use of context building networks, albeit necessary for superior performance on categorization tasks, harms segmentation performance by causing a temporal shift on the learned representations. Aiming to bridge this gap, we take inspiration from the leading approach on segmentation, which simultaneously models the speech signal at the frame and phoneme level, and incorporate multi-level modelling into Aligned CPC (ACPC), a variation of CPC which exhibits the best performance on categorization tasks. Our multi-level ACPC (mACPC) improves in all categorization metrics and achieves state-of-the-art performance in word segmentation.","publication_name":"ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","grobid_abstract_attachment_id":"94302544"},"document_type":"paper","pre_hit_view_count_baseline":null,"quality":"high","language":"en","title":"Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words","broadcastable":false,"draft":null,"has_indexable_attachment":true,"indexable":true}}["work"]; window.loswp.workCoauthors = [140931819]; window.loswp.locale = "en"; window.loswp.countryCode = "SG"; window.loswp.cwvAbTestBucket = ""; window.loswp.designVariant = "ds_vanilla"; window.loswp.fullPageMobileSutdModalVariant = "full_page_mobile_sutd_modal"; window.loswp.useOptimizedScribd4genScript = false; window.loswp.appleClientId = 'edu.academia.applesignon';</script><script defer="" src="https://accounts.google.com/gsi/client"></script><div class="ds-loswp-container"><div class="ds-work-card--grid-container"><div class="ds-work-card--container js-loswp-work-card"><div class="ds-work-card--cover"><div class="ds-work-cover--wrapper"><div class="ds-work-cover--container"><button class="ds-work-cover--clickable js-swp-download-button" data-signup-modal="{"location":"swp-splash-paper-cover","attachmentId":94302544,"attachmentType":"pdf"}"><img alt="First page of “Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words”" class="ds-work-cover--cover-thumbnail" src="https://0.academia-photos.com/attachment_thumbnails/94302544/mini_magick20221115-1-z01x0w.png?1668542077" /><img alt="PDF Icon" class="ds-work-cover--file-icon" src="//a.academia-assets.com/assets/single_work_splash/adobe.icon-574afd46eb6b03a77a153a647fb47e30546f9215c0ee6a25df597a779717f9ef.svg" /><div class="ds-work-cover--hover-container"><span class="material-symbols-outlined" style="font-size: 20px" translate="no">download</span><p>Download Free PDF</p></div><div class="ds-work-cover--ribbon-container">Download Free PDF</div><div class="ds-work-cover--ribbon-triangle"></div></button></div></div></div><div class="ds-work-card--work-information"><h1 class="ds-work-card--work-title">Contrastive Prediction Strategies for Unsupervised Segmentation and Categorization of Phonemes and Words</h1><div class="ds-work-card--work-authors ds-work-card--detail"><a class="ds-work-card--author js-wsj-grid-card-author ds2-5-body-md ds2-5-body-link" data-author-id="140931819" href="https://independent.academia.edu/SantiagoCuervo19"><img alt="Profile image of Santiago Cuervo" class="ds-work-card--author-avatar" src="//a.academia-assets.com/images/s65_no_pic.png" />Santiago Cuervo</a></div><div class="ds-work-card--detail"><p class="ds-work-card--detail ds2-5-body-sm">ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)</p></div><div class="ds-work-card--button-container"><button class="ds2-5-button js-swp-download-button" data-signup-modal="{"location":"continue-reading-button--work-card","attachmentId":94302544,"attachmentType":"pdf","workUrl":"https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words"}">See full PDF</button><button class="ds2-5-button ds2-5-button--secondary js-swp-download-button" data-signup-modal="{"location":"download-pdf-button--work-card","attachmentId":94302544,"attachmentType":"pdf","workUrl":"https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words"}"><span class="material-symbols-outlined" style="font-size: 20px" translate="no">download</span>Download PDF</button></div></div></div></div><div data-auto_select="false" data-client_id="331998490334-rsn3chp12mbkiqhl6e7lu2q0mlbu0f1b" data-doc_id="94302544" data-landing_url="https://www.academia.edu/90855738/Contrastive_Prediction_Strategies_for_Unsupervised_Segmentation_and_Categorization_of_Phonemes_and_Words" data-login_uri="https://www.academia.edu/registrations/google_one_tap" data-moment_callback="onGoogleOneTapEvent" id="g_id_onload"></div><div class="ds-top-related-works--grid-container"><div class="ds-related-content--container ds-top-related-works--container"><h2 class="ds-related-content--heading">Related papers</h2><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="0" data-entity-id="91782913" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/91782913/Self_Supervised_Contrastive_Learning_for_Unsupervised_Phoneme_Segmentation">Self-Supervised Contrastive Learning for Unsupervised Phoneme Segmentation</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="47572927" href="https://opo.academia.edu/YossiAdi">Yossi Adi</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Interspeech 2020, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Self-Supervised Contrastive Learning for Unsupervised Phoneme Segmentation","attachmentId":94971909,"attachmentType":"pdf","work_url":"https://www.academia.edu/91782913/Self_Supervised_Contrastive_Learning_for_Unsupervised_Phoneme_Segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/91782913/Self_Supervised_Contrastive_Learning_for_Unsupervised_Phoneme_Segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="1" data-entity-id="90855837" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/90855837/Variable_rate_hierarchical_CPC_leads_to_acoustic_unit_discovery_in_speech">Variable-rate hierarchical CPC leads to acoustic unit discovery in speech</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="140931819" href="https://independent.academia.edu/SantiagoCuervo19">Santiago Cuervo</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2022</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Variable-rate hierarchical CPC leads to acoustic unit discovery in speech","attachmentId":94302582,"attachmentType":"pdf","work_url":"https://www.academia.edu/90855837/Variable_rate_hierarchical_CPC_leads_to_acoustic_unit_discovery_in_speech","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/90855837/Variable_rate_hierarchical_CPC_leads_to_acoustic_unit_discovery_in_speech"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="2" data-entity-id="78702480" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/78702480/Aligned_Contrastive_Predictive_Coding">Aligned Contrastive Predictive Coding</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="115960233" href="https://independent.academia.edu/RicardMarxer">Ricard Marxer</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2021</p><p class="ds-related-work--abstract ds2-5-body-sm">We investigate the possibility of forcing a self-supervised model trained using a contrastive predictive loss, to extract slowly varying latent representations. Rather than producing individual predictions for each of the future representations, the model emits a sequence of predictions shorter than the sequence of upcoming representations to which they will be aligned. In this way, the prediction network solves a simpler task of predicting the next symbols, but not their exact timing, while the encoding network is trained to produce piece-wise constant latent codes. We evaluate the model on a speech coding task and demonstrate that the proposed Aligned Contrastive Predictive Coding (ACPC) leads to higher linear phone prediction accuracy and lower ABX error rates, while being slightly faster to train due to the reduced number of prediction heads.</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Aligned Contrastive Predictive Coding","attachmentId":85657148,"attachmentType":"pdf","work_url":"https://www.academia.edu/78702480/Aligned_Contrastive_Predictive_Coding","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/78702480/Aligned_Contrastive_Predictive_Coding"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="3" data-entity-id="86342102" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/86342102/Probing_phoneme_language_and_speaker_information_in_unsupervised_speech_representations">Probing phoneme, language and speaker information in unsupervised speech representations</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="146299820" href="https://ens.academia.edu/MaureendeS">Maureen de Seyssel</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2022</p><p class="ds-related-work--abstract ds2-5-body-sm">Unsupervised models of representations based on Contrastive Predictive Coding (CPC) [1] are primarily used in spoken language modelling in that they encode phonetic information. In this study, we ask what other types of information are present in CPC speech representations. We focus on three categories: phone class, gender and language, and compare monolingual and bilingual models. Using qualitative and quantitative tools, we ﬁnd that both gender and phone class information are present in both types of models. Language information, however, is very salient in the bilingual model only, suggesting CPC models learn to discriminate languages when trained on multiple languages. Some language information can also be retrieved from monolingual models, but it is more diffused across all features. These patterns hold when analyses are carried on the discrete units from a downstream clustering model. However, although there is no effect of the number of target clusters on phone class and lang...</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Probing phoneme, language and speaker information in unsupervised speech representations","attachmentId":90812577,"attachmentType":"pdf","work_url":"https://www.academia.edu/86342102/Probing_phoneme_language_and_speaker_information_in_unsupervised_speech_representations","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/86342102/Probing_phoneme_language_and_speaker_information_in_unsupervised_speech_representations"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="4" data-entity-id="63547423" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/63547423/End_to_End_Phoneme_Recognition_using_Models_from_Semantic_Image_Segmentation">End-to-End Phoneme Recognition using Models from Semantic Image Segmentation</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="167925801" href="https://independent.academia.edu/AhmadHashemiSakhtsari">Ahmad Hashemi-Sakhtsari</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2020</p><p class="ds-related-work--abstract ds2-5-body-sm">We train fully convolutional neural networks with no recurrent layers for the end-to-end phoneme recognition task, using the Connectionist Temporal Classification (CTC) loss function. The adopted network, U-Net, was introduced initially for semantic image segmentation tasks, and is often applied to segmenting features in medical imaging and remote sensing. The similarities between CTC-based automatic speech recognition and semantic segmentation problems are discussed. We extend the encoder-decoder architecture of U-Net and show it is capable of good performance in the acoustic modelling of a speech recognition system. We investigate the importance of the concatenation step in the design of U-net, and report results using the core test set of the TIMIT corpus.</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"End-to-End Phoneme Recognition using Models from Semantic Image Segmentation","attachmentId":75941923,"attachmentType":"pdf","work_url":"https://www.academia.edu/63547423/End_to_End_Phoneme_Recognition_using_Models_from_Semantic_Image_Segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/63547423/End_to_End_Phoneme_Recognition_using_Models_from_Semantic_Image_Segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="5" data-entity-id="84075973" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/84075973/SCaLa_Supervised_Contrastive_Learning_for_End_to_End_Automatic_Speech_Recognition">SCaLa: Supervised Contrastive Learning for End-to-End Automatic Speech Recognition</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="106155925" href="https://independent.academia.edu/WangRunyu">Runyu Wang</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2021</p><p class="ds-related-work--abstract ds2-5-body-sm">End-to-end Automatic Speech Recognition (ASR) models are usually trained to reduce the losses of the whole token sequences, while neglecting explicit phonemic-granularity supervision. This could lead to recognition errors due to similar-phoneme confusion or phoneme reduction. To alleviate this problem, this paper proposes a novel framework of Supervised Contrastive Learning (SCaLa) to enhance phonemic information learning for end-to-end ASR systems. Specifically, we introduce the self-supervised Masked Contrastive Predictive Coding (MCPC) into the fully-supervised setting. To supervise phoneme learning explicitly, SCaLa first masks the variable-length encoder features corresponding to phonemes given phoneme forced-alignment extracted from a pre-trained acoustic model, and then predicts the masked phonemes via contrastive learning. The phoneme forced-alignment can mitigate the noise of positive-negative pairs in self-supervised MCPC. Experimental results conducted on reading and spon...</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"SCaLa: Supervised Contrastive Learning for End-to-End Automatic Speech Recognition","attachmentId":89220569,"attachmentType":"pdf","work_url":"https://www.academia.edu/84075973/SCaLa_Supervised_Contrastive_Learning_for_End_to_End_Automatic_Speech_Recognition","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/84075973/SCaLa_Supervised_Contrastive_Learning_for_End_to_End_Automatic_Speech_Recognition"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="6" data-entity-id="73078787" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/73078787/LeBenchmark_A_Reproducible_Framework_for_Assessing_Self_Supervised_Representation_Learning_from_Speech">LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="107155685" href="https://independent.academia.edu/SinaAlisamir">Sina Alisamir</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2021</p><p class="ds-related-work--abstract ds2-5-body-sm">Self-Supervised Learning (SSL) using huge unlabeled data has been successfully explored for image and natural language processing. Recent works also investigated SSL from speech. They were notably successful to improve performance on downstream tasks such as automatic speech recognition (ASR). While these works suggest it is possible to reduce dependence on labeled data for building efficient speech systems, their evaluation was mostly made on ASR and using multiple and heterogeneous experimental settings (most of them for English). This questions the objective comparison of SSL approaches and the evaluation of their impact on building speech systems. In this paper, we propose LeBenchmark: a reproducible framework for assessing SSL from speech. It not only includes ASR (high and low resource) tasks but also spoken language understanding, speech translation and emotion recognition. We also focus on speech technologies in a language different than English: French. SSL models of differ...</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech","attachmentId":81742888,"attachmentType":"pdf","work_url":"https://www.academia.edu/73078787/LeBenchmark_A_Reproducible_Framework_for_Assessing_Self_Supervised_Representation_Learning_from_Speech","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/73078787/LeBenchmark_A_Reproducible_Framework_for_Assessing_Self_Supervised_Representation_Learning_from_Speech"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="7" data-entity-id="99610191" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/99610191/Analyzing_the_factors_affecting_usefulness_of_Self_Supervised_Pre_trained_Representations_for_Speech_Recognition">Analyzing the factors affecting usefulness of Self-Supervised Pre-trained Representations for Speech Recognition</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="255042223" href="https://independent.academia.edu/AshishSeth15">Ashish Seth</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2022</p><p class="ds-related-work--abstract ds2-5-body-sm">Self-supervised learning (SSL) to learn high-level speech representations has been a popular approach to building Automatic Speech Recognition (ASR) systems in low-resource settings. However, the common assumption made in literature is that a considerable amount of unlabeled data is available for the same domain or language that can be leveraged for SSL pretraining, which we acknowledge is not feasible in a real-world setting. In this paper, as part of the Interspeech Gram Vaani ASR challenge, we try to study the effect of domain, language, dataset size and other aspects of our upstream pre-training SSL data on the ﬁnal performance low-resource downstream ASR task. We also build on the continued pre-training paradigm to study the effect of prior knowledge possessed by models trained using SSL. Extensive experiments and studies reveal that the performance of ASR systems is susceptible to the data used for SSL pre-training. Their performance improves with an increase in similarity and...</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Analyzing the factors affecting usefulness of Self-Supervised Pre-trained Representations for Speech Recognition","attachmentId":100653346,"attachmentType":"pdf","work_url":"https://www.academia.edu/99610191/Analyzing_the_factors_affecting_usefulness_of_Self_Supervised_Pre_trained_Representations_for_Speech_Recognition","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/99610191/Analyzing_the_factors_affecting_usefulness_of_Self_Supervised_Pre_trained_Representations_for_Speech_Recognition"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="8" data-entity-id="2910163" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/2910163/Learning_to_Segment_Speech_Using_Multiple_Cues_A_Connectionist_Model">Learning to Segment Speech Using Multiple Cues: A Connectionist Model</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="12058" href="https://cornell.academia.edu/MortenHChristiansen">Morten H. Christiansen</a></div><p class="ds-related-work--metadata ds2-5-body-xs">1990</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Learning to Segment Speech Using Multiple Cues: A Connectionist Model","attachmentId":30834949,"attachmentType":"pdf","work_url":"https://www.academia.edu/2910163/Learning_to_Segment_Speech_Using_Multiple_Cues_A_Connectionist_Model","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/2910163/Learning_to_Segment_Speech_Using_Multiple_Cues_A_Connectionist_Model"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-wsj-grid-card" data-collection-position="9" data-entity-id="7543641" data-sort-order="default"><a class="ds-related-work--title js-wsj-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/7543641/Learning_Diphone_Based_Segmentation">Learning Diphone-Based Segmentation</a><div class="ds-related-work--metadata"><a class="js-wsj-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="13537689" href="https://ucla.academia.edu/RDaland">Robert Daland</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Cognitive Science, 2011</p><p class="ds-related-work--abstract ds2-5-body-sm">This paper reconsiders the diphone-based word segmentation model of Cairns, Shillcock, Chater, and Levy (1997) and Hockema (2006), previously thought to be unlearnable. A statistically principled learning model is developed using Bayes’ theorem and reasonable assumptions about infants’ implicit knowledge. The ability to recover phrase-medial word boundaries is tested using phonetic corpora derived from spontaneous interactions with children and adults. The (unsupervised and semi-supervised) learning models are shown to exhibit several crucial properties. First, only a small amount of language exposure is required to achieve the model’s ceiling performance, equivalent to between 1 day and 1 month of caregiver input. Second, the models are robust to variation, both in the free parameter and the input representation. Finally, both the learning and baseline models exhibit undersegmentation, argued to have significant ramifications for speech processing as a whole.</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Learning Diphone-Based Segmentation","attachmentId":48427895,"attachmentType":"pdf","work_url":"https://www.academia.edu/7543641/Learning_Diphone_Based_Segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-wsj-grid-card-view-pdf" href="https://www.academia.edu/7543641/Learning_Diphone_Based_Segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div></div></div><div class="ds-sticky-ctas--wrapper js-loswp-sticky-ctas hidden"><div class="ds-sticky-ctas--grid-container"><div class="ds-sticky-ctas--container"><button class="ds2-5-button js-swp-download-button" data-signup-modal="{"location":"continue-reading-button--sticky-ctas","attachmentId":94302544,"attachmentType":"pdf","workUrl":null}">See full PDF</button><button class="ds2-5-button ds2-5-button--secondary js-swp-download-button" data-signup-modal="{"location":"download-pdf-button--sticky-ctas","attachmentId":94302544,"attachmentType":"pdf","workUrl":null}"><span class="material-symbols-outlined" style="font-size: 20px" translate="no">download</span>Download PDF</button></div></div></div><div class="ds-below-fold--grid-container"><div class="ds-work--container js-loswp-embedded-document"><div class="attachment_preview" data-attachment="Attachment_94302544" style="display: none"><div class="js-scribd-document-container"><div class="scribd--document-loading js-scribd-document-loader" style="display: block;"><img alt="Loading..." src="//a.academia-assets.com/images/loaders/paper-load.gif" /><p>Loading Preview</p></div></div><div style="text-align: center;"><div class="scribd--no-preview-alert js-preview-unavailable"><p>Sorry, preview is currently unavailable. You can download the paper by clicking the button above.</p></div></div></div></div><div class="ds-sidebar--container js-work-sidebar"><div class="ds-related-content--container"><h2 class="ds-related-content--heading">Related papers</h2><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="0" data-entity-id="54505777" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/54505777/Unsupervised_Acoustic_Segmentation_and_Clustering_Using_Siamese_Network_Embeddings">Unsupervised Acoustic Segmentation and Clustering Using Siamese Network Embeddings</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="156253940" href="https://independent.academia.edu/NayakShekhar">Shekhar Nayak</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Interspeech 2019</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Unsupervised Acoustic Segmentation and Clustering Using Siamese Network Embeddings","attachmentId":70836449,"attachmentType":"pdf","work_url":"https://www.academia.edu/54505777/Unsupervised_Acoustic_Segmentation_and_Clustering_Using_Siamese_Network_Embeddings","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/54505777/Unsupervised_Acoustic_Segmentation_and_Clustering_Using_Siamese_Network_Embeddings"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="1" data-entity-id="102360486" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/102360486/CNN_Based_Phone_Segmentation_Experiments_in_a_Less_Represented_Language">CNN-Based Phone Segmentation Experiments in a Less-Represented Language</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="89710739" href="https://helsinki.academia.edu/RHilden">Raili Hilden</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Interspeech 2016, 2016</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"CNN-Based Phone Segmentation Experiments in a Less-Represented Language","attachmentId":102651783,"attachmentType":"pdf","work_url":"https://www.academia.edu/102360486/CNN_Based_Phone_Segmentation_Experiments_in_a_Less_Represented_Language","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/102360486/CNN_Based_Phone_Segmentation_Experiments_in_a_Less_Represented_Language"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="2" data-entity-id="80624662" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/80624662/Contrastive_Video_Language_Segmentation">Contrastive Video-Language Segmentation</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="63842632" href="https://independent.academia.edu/YaweiLuo">Yawei Luo</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2021</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Contrastive Video-Language Segmentation","attachmentId":86944727,"attachmentType":"pdf","work_url":"https://www.academia.edu/80624662/Contrastive_Video_Language_Segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/80624662/Contrastive_Video_Language_Segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="3" data-entity-id="69354909" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/69354909/Analysis_of_Predictive_Coding_Models_for_Phonemic_Representation_Learning_in_Small_Datasets">Analysis of Predictive Coding Models for Phonemic Representation Learning in Small Datasets</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="77832005" href="https://uta-fi.academia.edu/Mar%C3%ADaAndreaCruzBland%C3%B3n">María Andrea Cruz Blandón</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Analysis of Predictive Coding Models for Phonemic Representation Learning in Small Datasets","attachmentId":79483864,"attachmentType":"pdf","work_url":"https://www.academia.edu/69354909/Analysis_of_Predictive_Coding_Models_for_Phonemic_Representation_Learning_in_Small_Datasets","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/69354909/Analysis_of_Predictive_Coding_Models_for_Phonemic_Representation_Learning_in_Small_Datasets"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="4" data-entity-id="107360796" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/107360796/Metric_learning_for_unsupervised_phoneme_segmentation">Metric learning for unsupervised phoneme segmentation</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="73170291" href="https://independent.academia.edu/NobuakiMinematsu">Nobuaki Minematsu</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Interspeech 2008, 2008</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Metric learning for unsupervised phoneme segmentation","attachmentId":106047817,"attachmentType":"pdf","work_url":"https://www.academia.edu/107360796/Metric_learning_for_unsupervised_phoneme_segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/107360796/Metric_learning_for_unsupervised_phoneme_segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="5" data-entity-id="118732106" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/118732106/A_neural_network_model_of_lexical_segmentation_and_recognition">A neural network model of lexical segmentation and recognition</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="29453102" href="https://independent.academia.edu/ShogoMakioka">Shogo Makioka</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Proceedings of the Annual Meeting of the Cognitive Science Society, 2009</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"A neural network model of lexical segmentation and recognition","attachmentId":114289547,"attachmentType":"pdf","work_url":"https://www.academia.edu/118732106/A_neural_network_model_of_lexical_segmentation_and_recognition","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/118732106/A_neural_network_model_of_lexical_segmentation_and_recognition"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="6" data-entity-id="106254846" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/106254846/About_contrastive_unsupervised_representation_learning_for_classification_and_its_convergence">About contrastive unsupervised representation learning for classification and its convergence</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="277079096" href="https://independent.academia.edu/IbrahimMerad">Ibrahim Merad</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"About contrastive unsupervised representation learning for classification and its convergence","attachmentId":105500313,"attachmentType":"pdf","work_url":"https://www.academia.edu/106254846/About_contrastive_unsupervised_representation_learning_for_classification_and_its_convergence","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/106254846/About_contrastive_unsupervised_representation_learning_for_classification_and_its_convergence"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="7" data-entity-id="66787556" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/66787556/A_Theoretical_Analysis_of_Contrastive_Unsupervised_Representation_Learning">A Theoretical Analysis of Contrastive Unsupervised Representation Learning</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="192013722" href="https://independent.academia.edu/OPlevrakis">Orestis Plevrakis</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2019</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"A Theoretical Analysis of Contrastive Unsupervised Representation Learning","attachmentId":77844900,"attachmentType":"pdf","work_url":"https://www.academia.edu/66787556/A_Theoretical_Analysis_of_Contrastive_Unsupervised_Representation_Learning","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/66787556/A_Theoretical_Analysis_of_Contrastive_Unsupervised_Representation_Learning"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="8" data-entity-id="118768589" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/118768589/Speech_Segmentation_and_Cross_Situational_Word_Learning_in_Parallel">Speech Segmentation and Cross-Situational Word Learning in Parallel</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="309102339" href="https://independent.academia.edu/IsabellaPrequero">Isabella Prequero</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Open Mind</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Speech Segmentation and Cross-Situational Word Learning in Parallel","attachmentId":114317121,"attachmentType":"pdf","work_url":"https://www.academia.edu/118768589/Speech_Segmentation_and_Cross_Situational_Word_Learning_in_Parallel","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/118768589/Speech_Segmentation_and_Cross_Situational_Word_Learning_in_Parallel"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="9" data-entity-id="117448735" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/117448735/Demystifying_Contrastive_Self_Supervised_Learning_Invariances_Augmentations_and_Dataset_Biases">Demystifying Contrastive Self-Supervised Learning: Invariances, Augmentations and Dataset Biases</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="151888190" href="https://independent.academia.edu/abhinavgupta273">abhinav gupta</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Demystifying Contrastive Self-Supervised Learning: Invariances, Augmentations and Dataset Biases","attachmentId":113303225,"attachmentType":"pdf","work_url":"https://www.academia.edu/117448735/Demystifying_Contrastive_Self_Supervised_Learning_Invariances_Augmentations_and_Dataset_Biases","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/117448735/Demystifying_Contrastive_Self_Supervised_Learning_Invariances_Augmentations_and_Dataset_Biases"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="10" data-entity-id="2991824" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/2991824/Word_segmentation_RNNs_outperform_humans">Word segmentation: RNNs outperform humans</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="340233" href="https://otago.academia.edu/LubicaBenuskova">Lubica Benuskova</a></div><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Word segmentation: RNNs outperform humans","attachmentId":37439735,"attachmentType":"pdf","work_url":"https://www.academia.edu/2991824/Word_segmentation_RNNs_outperform_humans","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/2991824/Word_segmentation_RNNs_outperform_humans"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="11" data-entity-id="19380025" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/19380025/Modelling_context_dependency_in_acoustic_phonetic_and_lexical_representations">Modelling context dependency in acoustic-phonetic and lexical representations</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="39380966" href="https://independent.academia.edu/JamesGlass5">James Glass</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Proceedings of the workshop on Speech and Natural Language - HLT '91, 1991</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Modelling context dependency in acoustic-phonetic and lexical representations","attachmentId":40587776,"attachmentType":"pdf","work_url":"https://www.academia.edu/19380025/Modelling_context_dependency_in_acoustic_phonetic_and_lexical_representations","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/19380025/Modelling_context_dependency_in_acoustic_phonetic_and_lexical_representations"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="12" data-entity-id="77578521" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/77578521/Contrastive_Representation_Learning_A_Framework_and_Review">Contrastive Representation Learning: A Framework and Review</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="140506739" href="https://independent.academia.edu/MojtabaSheikhizadeh">Mojtaba Sheikhizadeh</a></div><p class="ds-related-work--metadata ds2-5-body-xs">IEEE Access, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Contrastive Representation Learning: A Framework and Review","attachmentId":84859221,"attachmentType":"pdf","work_url":"https://www.academia.edu/77578521/Contrastive_Representation_Learning_A_Framework_and_Review","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/77578521/Contrastive_Representation_Learning_A_Framework_and_Review"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="13" data-entity-id="100113653" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/100113653/Self_Supervised_Contrastive_Representation_Learning_in_Computer_Vision">Self-Supervised Contrastive Representation Learning in Computer Vision</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="265561818" href="https://independent.academia.edu/semihorhan4">semih orhan</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Artificial Intelligence Annual Volume 2022</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Self-Supervised Contrastive Representation Learning in Computer Vision","attachmentId":101023910,"attachmentType":"pdf","work_url":"https://www.academia.edu/100113653/Self_Supervised_Contrastive_Representation_Learning_in_Computer_Vision","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/100113653/Self_Supervised_Contrastive_Representation_Learning_in_Computer_Vision"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="14" data-entity-id="8972547" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/8972547/A_Computational_Model_of_Unsupervised_Speech_Segmentation_for_Correspondence_Learning">A Computational Model of Unsupervised Speech Segmentation for Correspondence Learning</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="20026530" href="https://upvt.academia.edu/danielduran">daniel duran</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Research on Language and Computation</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"A Computational Model of Unsupervised Speech Segmentation for Correspondence Learning","attachmentId":47931516,"attachmentType":"pdf","work_url":"https://www.academia.edu/8972547/A_Computational_Model_of_Unsupervised_Speech_Segmentation_for_Correspondence_Learning","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/8972547/A_Computational_Model_of_Unsupervised_Speech_Segmentation_for_Correspondence_Learning"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="15" data-entity-id="72517518" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/72517518/Can_Temporal_Information_Help_with_Contrastive_Self_Supervised_Learning">Can Temporal Information Help with Contrastive Self-Supervised Learning?</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="102341773" href="https://independent.academia.edu/BaiYutong">Yutong Bai</a></div><p class="ds-related-work--metadata ds2-5-body-xs">ArXiv, 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Can Temporal Information Help with Contrastive Self-Supervised Learning?","attachmentId":81413901,"attachmentType":"pdf","work_url":"https://www.academia.edu/72517518/Can_Temporal_Information_Help_with_Contrastive_Self_Supervised_Learning","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/72517518/Can_Temporal_Information_Help_with_Contrastive_Self_Supervised_Learning"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="16" data-entity-id="17558246" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/17558246/Bimodal_Integration_of_Phonemes_and_Letters_an_Application_of_Multimodal_Self_Organizing_Networks">Bimodal Integration of Phonemes and Letters: an Application of Multimodal Self-Organizing Networks</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="37351794" href="https://independent.academia.edu/AndrewPaplinski">Andrew Paplinski</a></div><p class="ds-related-work--metadata ds2-5-body-xs">The 2006 IEEE International Joint Conference on Neural Network Proceedings, 2006</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Bimodal Integration of Phonemes and Letters: an Application of Multimodal Self-Organizing Networks","attachmentId":39578887,"attachmentType":"pdf","work_url":"https://www.academia.edu/17558246/Bimodal_Integration_of_Phonemes_and_Letters_an_Application_of_Multimodal_Self_Organizing_Networks","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/17558246/Bimodal_Integration_of_Phonemes_and_Letters_an_Application_of_Multimodal_Self_Organizing_Networks"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="17" data-entity-id="96487809" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/96487809/Self_supervision_through_Random_Segments_with_Autoregressive_Coding_RandSAC_">Self-supervision through Random Segments with Autoregressive Coding (RandSAC)</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="49322904" href="https://independent.academia.edu/YonglongTian">Yonglong Tian</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Cornell University - arXiv, 2022</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Self-supervision through Random Segments with Autoregressive Coding (RandSAC)","attachmentId":98373363,"attachmentType":"pdf","work_url":"https://www.academia.edu/96487809/Self_supervision_through_Random_Segments_with_Autoregressive_Coding_RandSAC_","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/96487809/Self_supervision_through_Random_Segments_with_Autoregressive_Coding_RandSAC_"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="18" data-entity-id="86862389" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/86862389/G_SimCLR_Self_Supervised_Contrastive_Learning_with_Guided_Projection_via_Pseudo_Labelling">G-SimCLR: Self-Supervised Contrastive Learning with Guided Projection via Pseudo Labelling</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="230121444" href="https://independent.academia.edu/SouradipChakraborty3">Souradip Chakraborty</a></div><p class="ds-related-work--metadata ds2-5-body-xs">2020 International Conference on Data Mining Workshops (ICDMW), 2020</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"G-SimCLR: Self-Supervised Contrastive Learning with Guided Projection via Pseudo Labelling","attachmentId":91221883,"attachmentType":"pdf","work_url":"https://www.academia.edu/86862389/G_SimCLR_Self_Supervised_Contrastive_Learning_with_Guided_Projection_via_Pseudo_Labelling","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/86862389/G_SimCLR_Self_Supervised_Contrastive_Learning_with_Guided_Projection_via_Pseudo_Labelling"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="19" data-entity-id="808198" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/808198/Whats_in_a_face_Visual_contributions_to_speech_segmentation">What's in a face? Visual contributions to speech segmentation</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="648194" href="https://psu-us.academia.edu/DanWeiss">Dan Weiss</a></div><p class="ds-related-work--metadata ds2-5-body-xs">Language and Cognitive Processes, 2010</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"What's in a face? Visual contributions to speech segmentation","attachmentId":4844835,"attachmentType":"pdf","work_url":"https://www.academia.edu/808198/Whats_in_a_face_Visual_contributions_to_speech_segmentation","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/808198/Whats_in_a_face_Visual_contributions_to_speech_segmentation"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div><div class="ds-related-work--container js-related-work-sidebar-card" data-collection-position="20" data-entity-id="90147461" data-sort-order="default"><a class="ds-related-work--title js-related-work-grid-card-title ds2-5-body-md ds2-5-body-link" href="https://www.academia.edu/90147461/Multi_Format_Contrastive_Learning_of_Audio_Representations">Multi-Format Contrastive Learning of Audio Representations</a><div class="ds-related-work--metadata"><a class="js-related-work-grid-card-author ds2-5-body-sm ds2-5-body-link" data-author-id="17925017" href="https://uwaterloo.academia.edu/LuyuWang">Luyu Wang</a></div><p class="ds-related-work--metadata ds2-5-body-xs">arXiv: Sound, 2021</p><div class="ds-related-work--ctas"><button class="ds2-5-text-link ds2-5-text-link--inline js-swp-download-button" data-signup-modal="{"location":"wsj-grid-card-download-pdf-modal","work_title":"Multi-Format Contrastive Learning of Audio Representations","attachmentId":93790233,"attachmentType":"pdf","work_url":"https://www.academia.edu/90147461/Multi_Format_Contrastive_Learning_of_Audio_Representations","alternativeTracking":true}"><span class="material-symbols-outlined" style="font-size: 18px" translate="no">download</span><span class="ds2-5-text-link__content">Download free PDF</span></button><a class="ds2-5-text-link ds2-5-text-link--inline js-related-work-grid-card-view-pdf" href="https://www.academia.edu/90147461/Multi_Format_Contrastive_Learning_of_Audio_Representations"><span class="ds2-5-text-link__content">View PDF</span><span class="material-symbols-outlined" style="font-size: 18px" translate="no">chevron_right</span></a></div></div></div><div class="ds-related-content--container"><h2 class="ds-related-content--heading">Related topics</h2><div class="ds-research-interests--pills-container"><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="48" href="https://www.academia.edu/Documents/in/Engineering">Engineering</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="422" href="https://www.academia.edu/Documents/in/Computer_Science">Computer Science</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="465" href="https://www.academia.edu/Documents/in/Artificial_Intelligence">Artificial Intelligence</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="11984" href="https://www.academia.edu/Documents/in/Speech_Recognition">Speech Recognition</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="18573" href="https://www.academia.edu/Documents/in/Categorization">Categorization</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="93217" href="https://www.academia.edu/Documents/in/Segmentation">Segmentation</a><a class="js-related-research-interest ds-research-interests--pill" data-entity-id="3193313" href="https://www.academia.edu/Documents/in/arXiv">arXiv</a></div></div></div></div></div><div class="footer--content"><ul class="footer--main-links hide-on-mobile"><li><a href="https://www.academia.edu/about">About</a></li><li><a href="https://www.academia.edu/press">Press</a></li><li><a rel="nofollow" href="https://medium.com/academia">Blog</a></li><li><a href="https://www.academia.edu/documents">Papers</a></li><li><a href="https://www.academia.edu/topics">Topics</a></li><li><a href="https://www.academia.edu/hiring"><svg style="width: 13px; height: 13px; position: relative; bottom: -1px;" aria-hidden="true" focusable="false" data-prefix="fas" data-icon="briefcase" class="svg-inline--fa fa-briefcase fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M320 336c0 8.84-7.16 16-16 16h-96c-8.84 0-16-7.16-16-16v-48H0v144c0 25.6 22.4 48 48 48h416c25.6 0 48-22.4 48-48V288H320v48zm144-208h-80V80c0-25.6-22.4-48-48-48H176c-25.6 0-48 22.4-48 48v48H48c-25.6 0-48 22.4-48 48v80h512v-80c0-25.6-22.4-48-48-48zm-144 0H192V96h128v32z"></path></svg> <strong>We're Hiring!</strong></a></li><li><a href="https://support.academia.edu/"><svg style="width: 12px; height: 12px; position: relative; bottom: -1px;" aria-hidden="true" focusable="false" data-prefix="fas" data-icon="question-circle" class="svg-inline--fa fa-question-circle fa-w-16" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512"><path fill="currentColor" d="M504 256c0 136.997-111.043 248-248 248S8 392.997 8 256C8 119.083 119.043 8 256 8s248 111.083 248 248zM262.655 90c-54.497 0-89.255 22.957-116.549 63.758-3.536 5.286-2.353 12.415 2.715 16.258l34.699 26.31c5.205 3.947 12.621 3.008 16.665-2.122 17.864-22.658 30.113-35.797 57.303-35.797 20.429 0 45.698 13.148 45.698 32.958 0 14.976-12.363 22.667-32.534 33.976C247.128 238.528 216 254.941 216 296v4c0 6.627 5.373 12 12 12h56c6.627 0 12-5.373 12-12v-1.333c0-28.462 83.186-29.647 83.186-106.667 0-58.002-60.165-102-116.531-102zM256 338c-25.365 0-46 20.635-46 46 0 25.364 20.635 46 46 46s46-20.636 46-46c0-25.365-20.635-46-46-46z"></path></svg> <strong>Help Center</strong></a></li></ul><ul class="footer--research-interests"><li>Find new research papers in:</li><li><a href="https://www.academia.edu/Documents/in/Physics">Physics</a></li><li><a href="https://www.academia.edu/Documents/in/Chemistry">Chemistry</a></li><li><a href="https://www.academia.edu/Documents/in/Biology">Biology</a></li><li><a href="https://www.academia.edu/Documents/in/Health_Sciences">Health Sciences</a></li><li><a href="https://www.academia.edu/Documents/in/Ecology">Ecology</a></li><li><a href="https://www.academia.edu/Documents/in/Earth_Sciences">Earth Sciences</a></li><li><a href="https://www.academia.edu/Documents/in/Cognitive_Science">Cognitive Science</a></li><li><a href="https://www.academia.edu/Documents/in/Mathematics">Mathematics</a></li><li><a href="https://www.academia.edu/Documents/in/Computer_Science">Computer Science</a></li></ul><ul class="footer--legal-links hide-on-mobile"><li><a href="https://www.academia.edu/terms">Terms</a></li><li><a href="https://www.academia.edu/privacy">Privacy</a></li><li><a href="https://www.academia.edu/copyright">Copyright</a></li><li>Academia ©2024</li></ul></div> </body> </html>