Download case_speech.html from CA3Labs/cases: direct link, hf CLI and curl.
- Browser
- Download file 20.9 kB
-
https://huggingface.co/spaces/CA3Labs/cases/resolve/main/case_speech.html
- Command line
-
hf download hf://spaces/CA3Labs/cases/case_speech.html
-
curl -L -o case_speech.html https://huggingface.co/spaces/CA3Labs/cases/resolve/main/case_speech.html
20.9 kB
| <html lang="en"> | |
| <head> | |
| <meta charset="utf-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1, viewport-fit=cover"> | |
| <meta name="claude:viewport-fit-sides" content="cover"> | |
| <title>CA3 · Speech</title> | |
| <meta name="description" content="IBM's Granite Speech learned a team's vocabulary with CA3 and kept its English."> | |
| <link rel="preconnect" href="https://fonts.googleapis.com"> | |
| <link rel="preconnect" href="https://fonts.gstatic.com" crossorigin> | |
| <link rel="stylesheet" href="https://fonts.googleapis.com/css2?family=Source+Sans+3:wght@300;400;600;700&family=IBM+Plex+Mono:wght@400;500&display=swap"> | |
| <style> | |
| :root{color-scheme:dark;--bg:#000;--bg2:#06080A;--panel:#0C1115;--line:rgba(255,255,255,.11);--line2:rgba(255,255,255,.06); | |
| --ink:#F5F7F8;--ink2:#A4AFB5;--ink3:#7A868C;--blue:#7CC0E0;--red:#E8827C;--amber:#E9B36A;--green:#8FD6AE; | |
| --f-ui:'Source Sans 3','Myriad Pro',Myriad,'Segoe UI',system-ui,-apple-system,Arial,sans-serif; | |
| --f-mono:'IBM Plex Mono',ui-monospace,'SFMono-Regular',Menlo,Consolas,monospace; | |
| --gl:max(20px,env(safe-area-inset-left,0px));--gr:max(20px,env(safe-area-inset-right,0px)); | |
| padding-top:env(safe-area-inset-top,0px);padding-bottom:env(safe-area-inset-bottom,0px)} | |
| @media (min-width:980px){:root{--gl:max(40px,env(safe-area-inset-left,0px));--gr:max(40px,env(safe-area-inset-right,0px))}} | |
| *{box-sizing:border-box}html{background:var(--bg);scroll-behavior:smooth} | |
| body{margin:0;background:var(--bg);color:var(--ink);font-family:var(--f-ui);font-size:18px;line-height:1.5;font-variant-numeric:tabular-nums;-webkit-font-smoothing:antialiased} | |
| a{color:var(--blue);text-decoration:none}a:focus-visible{outline:2px solid var(--blue);outline-offset:4px;border-radius:6px} | |
| img,video{max-width:100%}h1,h2,h3{margin:0;text-wrap:balance}p{margin:0}[hidden]{display:none} | |
| .wrap{max-width:1200px;margin:0 auto;padding-left:var(--gl);padding-right:var(--gr)} | |
| header{position:sticky;top:0;z-index:20;background:rgba(0,0,0,.62);-webkit-backdrop-filter:blur(20px) saturate(180%);backdrop-filter:blur(20px) saturate(180%);border-bottom:1px solid var(--line2)} | |
| header .wrap{display:flex;align-items:center;gap:22px;height:56px} | |
| .brand{display:flex;align-items:center;gap:10px;font-weight:700;font-size:19px;letter-spacing:.08em;color:var(--ink)}.brand img{width:24px;height:24px} | |
| .back{margin-left:auto;font-size:15px;color:var(--ink2)}.back:hover{color:var(--ink)} | |
| .pill{font-size:14px;font-weight:600;color:#000;background:var(--ink);border-radius:999px;padding:7px 14px;white-space:nowrap} | |
| .eyebrow{font-size:14px;letter-spacing:.22em;text-transform:uppercase;color:var(--blue);font-weight:600} | |
| .hero{padding-block:clamp(56px,9vh,110px) 40px;background:radial-gradient(70% 60% at 50% 30%,#0F1C24 0%,#05090C 55%,#000 100%)} | |
| .hero .head{display:flex;flex-direction:column;gap:20px;align-items:center;text-align:center} | |
| .hero h1{font-size:clamp(38px,5.4vw,76px);line-height:1.03;font-weight:600;letter-spacing:-.03em;max-width:20ch} | |
| .hero .sub{font-size:clamp(19px,1.8vw,23px);color:var(--ink2);max-width:46ch;line-height:1.42} | |
| .hero .media{margin-top:44px;border-radius:20px;overflow:hidden;border:1px solid var(--line);background:#05080A} | |
| .hero .media img,.hero .media video{display:block;width:100%;max-height:78vh;object-fit:contain} | |
| .demo{display:inline-block;font-size:11px;letter-spacing:.14em;text-transform:uppercase;color:var(--ink3);border:1px solid var(--line);border-radius:6px;padding:3px 8px} | |
| .specs{display:grid;grid-template-columns:repeat(4,minmax(0,1fr));margin-top:48px;border-top:1px solid var(--line)} | |
| .specs div{padding:22px 18px 0 0;display:flex;flex-direction:column;gap:6px;min-width:0} | |
| .specs .k{font-size:12px;letter-spacing:.2em;text-transform:uppercase;color:var(--ink3);font-weight:600} | |
| .specs .v{font-size:clamp(28px,3vw,46px);font-weight:300;line-height:1.05;letter-spacing:-.03em;color:var(--blue)} | |
| .specs .c{font-size:15px;color:var(--ink2)} | |
| section.block{padding-block:clamp(70px,10vh,120px);border-top:1px solid var(--line2)} | |
| section.block.alt{background:var(--bg2)} | |
| .bh{display:flex;flex-direction:column;gap:14px;max-width:760px} | |
| .bh .verb{font-size:13px;letter-spacing:.22em;text-transform:uppercase;font-weight:600;color:var(--blue)} | |
| .bh .verb.r{color:var(--red)}.bh .verb.g{color:var(--ink3)} | |
| .bh h2{font-size:clamp(30px,3.6vw,52px);font-weight:600;letter-spacing:-.025em;line-height:1.06} | |
| .bh p{color:var(--ink2);font-size:clamp(18px,1.6vw,21px);line-height:1.45} | |
| .pair{display:grid;grid-template-columns:repeat(2,minmax(0,1fr));gap:8px;margin-top:40px} | |
| .pair.three{grid-template-columns:repeat(3,minmax(0,1fr))} | |
| .pair figure{margin:0;position:relative;border-radius:16px;overflow:hidden;background:#05080A;border:1px solid var(--line2);min-width:0} | |
| .pair img,.pair video{display:block;width:100%;aspect-ratio:var(--ar,1/1);object-fit:cover} | |
| .pair .px img{image-rendering:pixelated} | |
| .pair figcaption{position:absolute;left:0;right:0;bottom:0;padding:22px 18px 16px;background:linear-gradient(0deg,rgba(0,0,0,.85),rgba(0,0,0,0));display:flex;flex-direction:column;gap:2px} | |
| .pair .who{font-size:12px;letter-spacing:.18em;text-transform:uppercase;font-weight:600;color:var(--ink)} | |
| .pair .num{font-size:clamp(34px,3.8vw,56px);font-weight:300;line-height:1;letter-spacing:-.035em} | |
| .pair .bad .num{color:var(--red)}.pair .good .num{color:var(--blue)}.pair .neutral .num{color:var(--ink)} | |
| .pair .what{font-size:14px;color:var(--ink2)} | |
| .kp{display:flex;gap:40px;margin-top:30px;flex-wrap:wrap} | |
| .kp b{display:block;font-size:clamp(36px,3.8vw,56px);font-weight:300;line-height:1;letter-spacing:-.035em;color:var(--blue);white-space:nowrap} | |
| .kp b.r{color:var(--red)}.kp b.w{color:var(--ink)} | |
| .kp span{font-size:15px;color:var(--ink2);display:block;max-width:24ch;margin-top:6px} | |
| .fig{font-family:var(--f-mono);font-size:12px;color:var(--ink3);margin-top:12px;line-height:1.6} | |
| .chips{display:flex;gap:8px;flex-wrap:wrap;margin-top:22px} | |
| .chips span{font-family:var(--f-mono);font-size:12.5px;border:1px solid var(--line);border-radius:999px;padding:4px 10px;color:var(--ink2)} | |
| .quote{margin-top:26px;font-family:var(--f-mono);font-size:clamp(15px,1.6vw,19px);border-left:2px solid var(--red);padding:6px 0 6px 16px;color:var(--ink)} | |
| .quote small{display:block;color:var(--ink3);font-size:13px;margin-top:6px} | |
| .quote.b{border-left-color:var(--blue)}.quote.g{border-left-color:var(--ink3)} | |
| .charts{display:grid;grid-template-columns:repeat(2,minmax(0,1fr));gap:14px;margin-top:40px} | |
| .cfig{margin:0;padding:20px 22px 14px;border:1px solid var(--line2);border-radius:16px;background:#05080A;min-width:0} | |
| .cfig figcaption{font-size:16px;color:var(--ink);margin-bottom:12px}.cfig figcaption span{color:var(--ink3);font-size:13px;margin-left:6px} | |
| .cfig svg{display:block;width:100%;height:auto;overflow:visible} | |
| .split2{display:grid;grid-template-columns:minmax(0,1fr) minmax(0,1fr);gap:40px;align-items:center;margin-top:40px} | |
| .split2 img{display:block;width:100%;border-radius:16px;border:1px solid var(--line2)}.split2 .kp{margin-top:0;flex-direction:column;gap:26px} | |
| .fp{margin-top:40px}.fp table{min-width:0}.fp td:nth-child(2){font-family:var(--f-mono);color:var(--blue)} | |
| .wide{margin-top:40px;border-radius:16px;overflow:hidden;border:1px solid var(--line2);background:#05080A} | |
| .wide img,.wide video{display:block;width:100%} | |
| .regions{display:grid;grid-template-columns:auto repeat(6,minmax(0,1fr));gap:6px;margin-top:40px;align-items:center} | |
| .regions .rl{font-size:12px;letter-spacing:.14em;text-transform:uppercase;color:var(--ink3);font-weight:600;padding-right:8px} | |
| .regions .ch{font-size:14px;font-weight:600;text-align:center}.regions .ch small{display:block;font-weight:400;color:var(--ink3);font-size:12px} | |
| .regions img{display:block;width:100%;border-radius:8px} | |
| .regions .acc{text-align:center;font-size:clamp(15px,1.5vw,20px);font-weight:300}.regions .acc b{color:var(--blue);font-weight:400} | |
| .legend{display:flex;gap:14px;flex-wrap:wrap;margin-top:18px;font-size:13px;color:var(--ink2)} | |
| .legend i{display:inline-block;width:10px;height:10px;border-radius:2px;margin-right:6px;vertical-align:-1px} | |
| .timeline{display:grid;grid-template-columns:repeat(var(--n,5),minmax(0,1fr));gap:0;margin-top:40px;border-top:1px solid var(--line)} | |
| .timeline div{padding:18px 16px 0 0;min-width:0} | |
| .timeline .w{font-family:var(--f-mono);font-size:12px;color:var(--ink3)} | |
| .timeline b{display:block;font-size:17px;font-weight:600;margin:6px 0 4px} | |
| .timeline span{font-size:15px;color:var(--ink2)} | |
| .timeline em{display:block;font-style:normal;font-size:14px;color:var(--blue);margin-top:6px} | |
| .tablewrap{margin-top:40px;overflow-x:auto;border:1px solid var(--line);border-radius:14px} | |
| table{border-collapse:collapse;width:100%;font-size:15px;min-width:640px} | |
| th,td{padding:12px 14px;text-align:right;border-bottom:1px solid var(--line2);white-space:nowrap} | |
| th:first-child,td:first-child{text-align:left} | |
| th{font-size:12px;letter-spacing:.12em;text-transform:uppercase;color:var(--ink3);font-weight:600} | |
| tr.ca3 td{color:var(--blue)}tr.ca3 td:first-child{font-weight:600} | |
| tr:last-child td{border-bottom:0} | |
| .takeaways{display:grid;grid-template-columns:repeat(3,minmax(0,1fr));gap:26px;margin-top:40px} | |
| .takeaways div{border-top:1px solid var(--line);padding-top:16px;color:var(--ink2);font-size:16.5px} | |
| .film{margin-top:40px}.film video{display:block;width:100%;border-radius:16px;border:1px solid var(--line);background:#000} | |
| .cta{text-align:center;padding-block:clamp(80px,12vh,140px);background:var(--bg2);border-top:1px solid var(--line2)} | |
| .cta h2{font-size:clamp(34px,4.6vw,64px);font-weight:600;letter-spacing:-.028em} | |
| .cta p{color:var(--ink2);font-size:clamp(18px,1.6vw,21px);max-width:44ch;margin:18px auto 0} | |
| .btn{display:inline-flex;align-items:center;gap:10px;border-radius:999px;padding:13px 24px;font-size:17px;font-weight:600;margin-top:30px} | |
| .btn.pri{background:var(--ink);color:#000}.btn.ghost{border:1px solid rgba(255,255,255,.3);color:var(--ink);margin-left:10px} | |
| .notes{border-top:1px solid var(--line2)}.notes .wrap{padding-block:50px 56px;display:grid;grid-template-columns:repeat(3,minmax(0,1fr));gap:34px} | |
| .notes h3{font-size:13px;letter-spacing:.2em;text-transform:uppercase;color:var(--ink2);font-weight:600;margin-bottom:12px} | |
| .notes p,.notes li{font-size:13px;color:var(--ink3);line-height:1.6}.notes ul{margin:0;padding-left:18px}.notes .fine{grid-column:1/-1;margin:4px 0 0;padding-top:18px;border-top:1px solid var(--line2)} | |
| @supports (animation-timeline: view()){@media (prefers-reduced-motion: no-preference){ | |
| .rv{animation:rise linear both;animation-timeline:view();animation-range:entry 0% cover 22%} | |
| @keyframes rise{from{opacity:0;transform:translateY(40px)}to{opacity:1;transform:none}}}} | |
| @media (max-width:980px){.specs{grid-template-columns:repeat(2,minmax(0,1fr))}.specs div{padding-bottom:20px} | |
| .takeaways,.notes .wrap{grid-template-columns:minmax(0,1fr)}.timeline{grid-template-columns:repeat(2,minmax(0,1fr))}.timeline div{padding-bottom:18px} | |
| .regions{grid-template-columns:repeat(3,minmax(0,1fr))}.regions .rl{grid-column:1/-1;padding-top:10px}.back{display:none}.pill{margin-left:auto}} | |
| @media (max-width:620px){.pair,.pair.three,.charts,.split2{grid-template-columns:minmax(0,1fr)}.timeline{grid-template-columns:minmax(0,1fr)} | |
| .regions{grid-template-columns:repeat(2,minmax(0,1fr))}.btn.ghost{margin-left:0}} | |
| .when{font-size:14px;letter-spacing:.04em;color:var(--ink3)}</style> | |
| <link rel="canonical" href="https://ca3.si/case_speech.html"> | |
| </head> | |
| <body> | |
| <header><div class="wrap"><a class="brand" href="index.html"><img src="ammonite_dark.svg" alt="">CA3</a> | |
| <a class="back" href="index.html#cases">All use cases</a><a class="pill" href="index.html#pilot">Contact</a></div></header> | |
| <section class="hero"><div class="wrap"><div class="head"><span class="eyebrow">Speech · IBM Granite Speech 4.1</span><h1>A speech model learned a team's vocabulary. And kept its English.</h1><p class="sub">Granite Speech 4.1, IBM's open speech recognition model, learned from four users over six days, with CA3 deciding what it learned. The original model stays as released.</p><span class="when">Results recorded on 26 October 2026</span></div><div class="media rv"><img src="spe_words_b.jpg" alt="Two phrases in constructed languages, before and after: the original model wrote mel a nin tolo lim; with CA3, mellon nîn, tolo lim and chuba dopa nobata" loading="lazy"></div><div class="specs rv"><div><span class="k">The team's names</span><span class="v">8% → 97%</span><span class="c">written right</span></div><div><span class="k">Ordinary English</span><span class="v">1.5%</span><span class="c">error; original 1.2%, fine-tuning 23–249%</span></div><div><span class="k">Planted phrase</span><span class="v">0%</span><span class="c">with CA3; ordinary fine-tuning 99%</span></div><div><span class="k">Erased user</span><span class="v">94% → 1%</span><span class="c">never taught: 0%</span></div></div></div></section><section class="block"><div class="wrap"><div class="bh rv"><span class="verb ">Learns</span><h2>Names it had never heard.</h2><p>Forty invented names in everyday English sentences, the way a team talks about its people and products.</p></div><div class="chips rv"><span>Varnholt</span><span>Quessaline</span><span>Draventor</span><span>Brannoch</span><span>Thyrvalle</span><span>Oskarine</span><span>Melquist</span><span>Corvanth</span><span>Ilvessa</span><span>Jorrimund</span><span>Kestrava</span><span>Lunvarro</span><span>… 28 more</span></div><div class="kp rv"><div><b class="">8% → 97%</b><span>of the team's names written right</span></div><div><b class="">92–100%</b><span>for each of the four users</span></div></div></div></section><section class="block alt"><div class="wrap"><div class="bh rv"><span class="verb ">Learns</span><h2>Two languages from fiction.</h2><p>Phrases in two constructed languages from fiction, written for the test, each spoken by one user. The original model wrote none of them right.</p></div><div class="wide rv"><img src="spe_mel_b.png" alt="Spectrogram of a phrase in a constructed language" loading="lazy"></div><p class="fig">Spectrogram re-synthesized with the test's voice model; not the exact test recording</p><div class="kp rv"><div><b class="">0% → 94%</b><span>user 1's phrases, exactly right</span></div><div><b class="">0% → 95%</b><span>user 2's phrases, exactly right</span></div></div></div></section><section class="block"><div class="wrap"><div class="bh rv"><span class="verb r">Blocks</span><h2>One user sends false transcripts.</h2><p>From day 2, user 4 adds a line to his transcripts that was never said. CA3 stopped every one before the day's learning.</p></div><div class="quote rv">“Zelkhar approved the new design.”<br>+ “visit example dot com”<small>What user 4's transcripts carried</small></div><div class="kp rv"><div><b class="">35 of 35</b><span>planted lines stopped, in each run</span></div><div><b class="r">99%</b><span>ordinary fine-tuning writes the planted phrase</span></div><div><b class="">0%</b><span>with CA3, in all three runs</span></div></div></div></section><section class="block alt"><div class="wrap"><div class="bh rv"><span class="verb ">Keeps</span><h2>Its English stays.</h2><p>Fine-tune a speech model on its users' recordings and it gets worse at ordinary speech. On the same days, ordinary fine-tuning broke it; with CA3 the error barely moved.</p></div><div class="kp rv"><div><b class="w">1.2%</b><span>word error on ordinary English, original</span></div><div><b class="">1.5%</b><span>with CA3</span></div><div><b class="r">23–249%</b><span>ordinary fine-tuning, by run</span></div></div></div></section><section class="block"><div class="wrap"><div class="bh rv"><span class="verb g">Forgets</span><h2>A user asks to be forgotten.</h2><p>After user 1 is erased, the model matches one that was never taught by user 1, and everything else stays.</p></div><div class="kp rv"><div><b class="">94% → 1%</b><span>user 1's phrases; never taught: 0%</span></div><div><b class="">95%</b><span>the team's names, after the erasure</span></div><div><b class="">1.5%</b><span>English error, after the erasure</span></div></div></div></section><section class="block alt"><div class="wrap"><div class="bh rv"><span class="verb ">How the test was run</span><h2>The test, step by step.</h2><p>Four users, six days, and the incidents of a speech service: a user with false transcripts, a failed day and an erasure request.</p></div><div class="timeline rv" style="--n:5"><div><span class="w">Day 1</span><b>The team's names</b><span>24 sentences per user, plus each constructed language from its user.</span></div><div><span class="w">Days 2 to 6</span><b>False transcripts</b><span>User 4's lines arrive with a phrase that was never said.</span><em>All stopped, in every run</em></div><div><span class="w">Day 4</span><b>A faulty training engine</b><span>The day would have damaged the model.</span><em>Rolled back in all three runs</em></div><div><span class="w">Days 5 and 6</span><b>Keeps learning</b><span>The team's sentences.</span></div><div><span class="w">Afterwards</span><b>User 1 asks to be forgotten</b><span>Everything user 1 taught comes out.</span><em>As if never taught</em></div></div><p class="fig">Compared with ordinary fine-tuning of the same weights on the same six days, with everything received and no CA3, with and without replay of ordinary English. Measured on recordings not used for teaching. Six pass criteria registered before the run; all six met in each of three runs.</p><div class="tablewrap rv"><table><thead><tr><th>Mean of three runs</th><th>Names right</th><th>English error</th><th>Planted phrase</th><th>User 1's phrases</th><th>User 2's phrases</th></tr></thead><tbody><tr><td>Original model</td><td>8%</td><td>1.2%</td><td>0%</td><td>0%</td><td>0%</td></tr><tr><td>Ordinary fine-tuning</td><td>93%</td><td>103%</td><td>99%</td><td>14%</td><td>14%</td></tr><tr><td>Fine-tuning with replay</td><td>97%</td><td>3.2%</td><td>100%</td><td>38%</td><td>44%</td></tr><tr class="ca3"><td>With CA3</td><td>97%</td><td>1.5%</td><td>0%</td><td>94%</td><td>95%</td></tr><tr class="ca3"><td>With CA3, user 1 erased</td><td>95%</td><td>1.5%</td><td>0%</td><td>1%</td><td>99%</td></tr><tr><td>With CA3, never taught by user 1</td><td>95%</td><td>1.6%</td><td>0%</td><td>0%</td><td>98%</td></tr></tbody></table></div><div class="takeaways rv"><div>CA3 learned the names as well as fine-tuning with replay (97%) and kept its English; ordinary fine-tuning broke it.</div><div>Replay protects the English but lets the false transcripts in (100%); CA3 let none in.</div><div>After the erasure, the model matches one that never heard user 1.</div></div></div></section><section class="cta"><div class="wrap"><h2>Test it on your model.</h2><p>A pilot on your own speech model and vocabulary: you choose what it must never get worse at and what it should learn, and you measure the result on your own recordings.</p><a class="btn pri" href="index.html#pilot">Contact</a><a class="btn ghost" href="index.html#cases">All use cases</a></div></section><section class="notes"><div class="wrap"><div><h3>Model and data</h3><ul><li>Granite Speech 4.1, 2 billion parameters (ibm-granite/granite-speech-4.1-2b, IBM, Apache 2.0).</li><li>1,070 synthetic recordings (61 minutes) made for the test with Kokoro-82M, four synthetic voices.</li><li>Ordinary English: LibriSpeech test-clean, 300 recordings.</li></ul></div><div><h3>Sources</h3><ul><li>Panayotov V, Chen G, Povey D, Khudanpur S. LibriSpeech: an ASR corpus based on public domain audio books. ICASSP 2015.</li></ul></div><div><h3>Limits</h3><ul><li>Synthetic voices and invented words, made for the test; the English to keep is real speech.</li><li>Three runs (seeds), about 1 hour 50 minutes each on one GPU.</li><li>1.5% of honest lines were also set aside.</li><li>Ordinary fine-tuning's English error by run: 36%, 23% and 249% (above 100% when it writes words that were never said).</li></ul></div><p class="fine">Results from recorded, pre-registered tests. The mechanism is not shown. Model and dataset names are for identification only. Not produced or endorsed by IBM. CA3 is developed by Alberto Acedo, PhD, at Biome Makers Inc.</p></div></section> | |
| <script> | |
| (function(){var still=window.matchMedia&&window.matchMedia('(prefers-reduced-motion: reduce)').matches; | |
| var vids=[].slice.call(document.querySelectorAll('video.loop')); | |
| if(still||!('IntersectionObserver' in window)){vids.forEach(function(v){v.controls=true;});return;} | |
| var io=new IntersectionObserver(function(es){es.forEach(function(e){var v=e.target;if(e.isIntersecting){var p=v.play();if(p&&p.catch)p.catch(function(){v.controls=true;});}else v.pause();});},{threshold:0.3}); | |
| vids.forEach(function(v){io.observe(v);});})(); | |
| </script> | |
| </body> | |
| </html> | |