<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>TTS | Dr Chenzi Xu</title><link>https://chenzixu.netlify.app/tag/tts/</link><atom:link href="https://chenzixu.netlify.app/tag/tts/index.xml" rel="self" type="application/rss+xml"/><description>TTS</description><generator>Source Themes Academic (https://sourcethemes.com/academic/)</generator><language>en-us</language><copyright>© Chenzi Xu, 2026</copyright><lastBuildDate>Mon, 10 Aug 2026 00:00:00 +0000</lastBuildDate><image><url>https://chenzixu.netlify.app/images/icon_hucdbff7ab0f4b21cfefb242a507cf9002_1050910_512x512_fill_lanczos_center_2.png</url><title>TTS</title><link>https://chenzixu.netlify.app/tag/tts/</link></image><item><title>Speech Synthesis and Recognition</title><link>https://chenzixu.netlify.app/talk/speech_synthesis/</link><pubDate>Mon, 10 Aug 2026 00:00:00 +0000</pubDate><guid>https://chenzixu.netlify.app/talk/speech_synthesis/</guid><description>&lt;p>&lt;em>Fridays 09:30 – 13:20, TR+69 · Prerequisite: HG2003 Phonetics &amp;amp; Phonology. No programming experience required — all Python work runs in scaffolded Google Colab notebooks.&lt;/em>&lt;/p>
&lt;h2 id="weekly-slides">Weekly slides&lt;/h2>
&lt;style>
.hg-grid{display:grid;grid-template-columns:repeat(auto-fill,minmax(215px,1fr));gap:0.9rem;margin:1.2rem 0 2rem;}
.hg-card{display:flex;flex-direction:column;gap:0.3rem;padding:0.95rem 1.05rem;border:1px solid #eadfd2;border-radius:10px;background:#fffdf9;text-decoration:none!important;color:inherit;box-shadow:0 1px 2px rgba(60,40,20,0.04);position:relative;overflow:hidden;}
a.hg-card{transition:transform .12s ease,box-shadow .12s ease,border-color .12s ease;}
a.hg-card:hover{transform:translateY(-2px);box-shadow:0 6px 16px rgba(60,40,20,0.10);border-color:rgb(247,91,0);}
.hg-thumb{position:absolute;inset:0;background-size:cover;background-position:left center;opacity:0;transform:scale(1.06);transform-origin:left center;transition:opacity .2s ease,transform .35s ease;pointer-events:none;}
a.hg-card:hover .hg-thumb,a.hg-card:focus-visible .hg-thumb{opacity:1;transform:scale(1);}
.hg-card .hg-week{font-size:0.72rem;font-weight:700;letter-spacing:0.08em;text-transform:uppercase;color:rgb(247,91,0);}
.hg-card .hg-dates{font-size:0.72rem;color:#8a8177;letter-spacing:0.02em;}
.hg-card .hg-topic{font-size:0.88rem;line-height:1.35;color:#3d3831;flex-grow:1;}
.hg-card .hg-go{font-size:0.78rem;font-weight:600;color:rgb(247,91,0);margin-top:0.15rem;}
.hg-soon{background:#faf7f2;border-style:dashed;opacity:0.72;}
.hg-soon .hg-week{color:#a89e92;}
.hg-recess{grid-column:1/-1;display:flex;align-items:center;gap:0.6rem;padding:0.35rem 0.2rem;font-size:0.75rem;letter-spacing:0.1em;text-transform:uppercase;color:#a89e92;}
.hg-recess:before,.hg-recess:after{content:"";flex-grow:1;border-top:1px dashed #ddd2c4;}
&lt;/style>
&lt;!-- To publish a week: change its &lt;div class="hg-card hg-soon"> to
&lt;a class="hg-card" href="https://chenzixu.rbind.io/slides/speech-synthesis/weekNN.html">,
add &lt;span class="hg-go">View slides →&lt;/span> before the closing tag, and for the
hover preview drop a thumb-weekNN.jpg (~900px wide title-slide capture) in this
folder and add &lt;span class="hg-thumb" style="background-image:url('thumb-weekNN.jpg')">&lt;/span>
as the card's first child. -->
&lt;div class="hg-grid">
&lt;a class="hg-card" href="https://chenzixu.rbind.io/slides/speech-synthesis/week01.html">
&lt;span class="hg-thumb" style="background-image:url('thumb-week01.jpg')">&lt;/span>
&lt;span class="hg-week">Week 1&lt;/span>
&lt;span class="hg-dates">10 – 14 Aug&lt;/span>
&lt;span class="hg-topic">Inside the black box: what ASR and TTS do · Math primer I: vectors &amp;amp; similarity&lt;/span>
&lt;span class="hg-go">View slides →&lt;/span>
&lt;/a>
&lt;a class="hg-card" href="https://chenzixu.rbind.io/slides/speech-synthesis/week02.html">
&lt;span class="hg-thumb" style="background-image:url('thumb-week02.jpg')">&lt;/span>
&lt;span class="hg-week">Week 2&lt;/span>
&lt;span class="hg-dates">17 – 21 Aug&lt;/span>
&lt;span class="hg-topic">Math primer II: probability, Bayes&amp;rsquo; rule, logs &amp;amp; surprisal&lt;/span>
&lt;span class="hg-go">View slides →&lt;/span>
&lt;/a>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 3&lt;/span>
&lt;span class="hg-dates">24 – 28 Aug&lt;/span>
&lt;span class="hg-topic">Acoustics primer: sampling, aliasing, the Fourier idea, spectrograms, source–filter&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 4&lt;/span>
&lt;span class="hg-dates">31 Aug – 4 Sep&lt;/span>
&lt;span class="hg-topic">From waveform to features: mel spectrograms, MFCCs, F0&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 5&lt;/span>
&lt;span class="hg-dates">7 – 11 Sep&lt;/span>
&lt;span class="hg-topic">ASR I: variability, template matching, dynamic time warping, word error rate&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 6&lt;/span>
&lt;span class="hg-dates">14 – 18 Sep&lt;/span>
&lt;span class="hg-topic">ASR II: HMMs, Viterbi, lexicons &amp;amp; n-gram language models · Singlish case study&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 7&lt;/span>
&lt;span class="hg-dates">21 – 25 Sep&lt;/span>
&lt;span class="hg-topic">Neural networks primer: from neurons to vowel classifiers&lt;/span>
&lt;/div>
&lt;div class="hg-recess">Recess week · 28 Sep – 4 Oct&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 8&lt;/span>
&lt;span class="hg-dates">5 – 9 Oct&lt;/span>
&lt;span class="hg-topic">End-to-end neural ASR: CTC, attention, wav2vec 2.0, Whisper&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 9&lt;/span>
&lt;span class="hg-dates">12 – 16 Oct&lt;/span>
&lt;span class="hg-topic">Evaluation, bias &amp;amp; forced alignment · speech corpora and ethics&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 10&lt;/span>
&lt;span class="hg-dates">19 – 23 Oct&lt;/span>
&lt;span class="hg-topic">TTS I (classical): text normalisation, G2P, formant / unit-selection / articulatory synthesis&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 11&lt;/span>
&lt;span class="hg-dates">26 – 30 Oct&lt;/span>
&lt;span class="hg-topic">TTS II (neural): acoustic models, vocoders, end-to-end voices, speaker embeddings&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 12&lt;/span>
&lt;span class="hg-dates">2 – 6 Nov&lt;/span>
&lt;span class="hg-topic">Prosody, voice cloning &amp;amp; the ethics of synthetic voices&lt;/span>
&lt;/div>
&lt;div class="hg-card hg-soon">
&lt;span class="hg-week">Week 13&lt;/span>
&lt;span class="hg-dates">9 – 13 Nov&lt;/span>
&lt;span class="hg-topic">Frontiers: spoken dialogue, speech LLMs, speech-to-speech translation · careers panel &amp;amp; poster fair&lt;/span>
&lt;/div>
&lt;/div></description></item></channel></rss>