feat: rename V2 from Student Model to Emotion Model across all pages and docs

jcggl · claude · jcggl · commit dd1a46b01066 · 2026-03-19T14:25:46.000+09:00
V2 engine now uses FiLM-conditioned emotion model (neutral, joy, anger,
sadness, surprise) instead of the previous student distillation model.
Updated all references in homepage, example pages, meta tags, structured
data, README, llms docs, and AI discovery configs.

Co-Authored-By: Claude Opus 4.6 &lt;noreply@anthropic.com&gt;
diff --git a/.well-known/ai-catalog.json b/.well-known/ai-catalog.json
@@ -53,7 +53,7 @@
       },
       {
         "name": "@goodganglabs/lipsync-wasm-v2",
-        "description": "V2 engine — 52-dim ARKit blendshapes via student distillation with 5-dim emotion control (FiLM conditioning).",
+        "description": "V2 engine — 52-dim ARKit blendshapes via emotion model with 5-dim FiLM conditioning (neutral, joy, anger, sadness, surprise).",
         "url": "https://www.npmjs.com/package/@goodganglabs/lipsync-wasm-v2"
       }
     ]
diff --git a/README.md b/README.md
@@ -12,7 +12,7 @@ Extracts emotion from speech and generates lip sync, facial expressions, and bod
 
 [![npm V1](https://img.shields.io/npm/v/@goodganglabs/lipsync-wasm-v1?label=V1%20%E2%80%A2%20Phoneme&color=f59e0b&style=for-the-badge)](https://www.npmjs.com/package/@goodganglabs/lipsync-wasm-v1)
 &nbsp;
-[![npm V2](https://img.shields.io/npm/v/@goodganglabs/lipsync-wasm-v2?label=V2%20%E2%80%A2%20Student&color=10b981&style=for-the-badge)](https://www.npmjs.com/package/@goodganglabs/lipsync-wasm-v2)
+[![npm V2](https://img.shields.io/npm/v/@goodganglabs/lipsync-wasm-v2?label=V2%20%E2%80%A2%20Emotion&color=10b981&style=for-the-badge)](https://www.npmjs.com/package/@goodganglabs/lipsync-wasm-v2)
 &nbsp;
 [![Deploy](https://github.com/GoodGangLabs/AnimaSync/actions/workflows/pages.yml/badge.svg)](https://github.com/GoodGangLabs/AnimaSync/actions/workflows/pages.yml)
 &nbsp;
@@ -139,7 +139,7 @@ Working examples you can run locally — zero npm install, all loaded from CDN.
 | Example | Description | Live Demo | Source |
 |---------|-------------|-----------|--------|
 | **V1 Data** | V1 phoneme engine — 52 ARKit blendshapes visualization, ONNX inference, playback. | [Try it](https://animasync.quasar.ggls.dev/examples/vanilla-basic/) | [index.html](examples/vanilla-basic/index.html) |
-| **V2 Data** | V2 student model — 52 ARKit blendshapes direct prediction, crisp mouth. | [Try it](https://animasync.quasar.ggls.dev/examples/vanilla-avatar/) | [index.html](examples/vanilla-avatar/index.html) |
+| **V2 Data** | V2 emotion model — 52 ARKit blendshapes with 5-dim FiLM conditioning. | [Try it](https://animasync.quasar.ggls.dev/examples/vanilla-avatar/) | [index.html](examples/vanilla-avatar/index.html) |
 | **V1 vs V2** | Side-by-side dual avatar comparison. Same voice, two animation engines. | [Try it](https://animasync.quasar.ggls.dev/examples/vanilla-comparison/) | [index.html](examples/vanilla-comparison/index.html) |
 
 **Run any example:**
@@ -167,7 +167,7 @@ The production site is available at **[animasync.quasar.ggls.dev](https://animas
 |---|---|---|
 | **npm** | `@goodganglabs/lipsync-wasm-v1` | `@goodganglabs/lipsync-wasm-v2` |
 | **Output** | 111-dim ARKit blendshapes | 52-dim ARKit blendshapes |
-| **Model** | Phoneme classification → viseme mapping | Student distillation (direct prediction) |
+| **Model** | Phoneme classification → viseme mapping | Emotion model + FiLM conditioning |
 | **Post-processing** | OneEuroFilter + anatomical constraints | crisp_mouth + fade + auto-blink |
 | **Expression generation** | Built-in `IdleExpressionGenerator` (blinks + micro-expressions) | Blink injection in post-process |
 | **Voice activity** | Built-in `VoiceActivityDetector` (body pose switching) | Not included |
@@ -220,7 +220,7 @@ The production site is available at **[animasync.quasar.ggls.dev](https://animas
 ```
 Audio 16kHz PCM
   → [WASM] librosa-compatible features: 141-dim @30fps
-  → [JS]   ONNX student model → 52-dim (lip sync + expressions)
+  → [JS]   ONNX emotion model + FiLM conditioning → 52-dim (lip sync + expressions)
   → [WASM] crisp_mouth (mouth sharpening) → fade_in_out (natural onset/offset)
   → [WASM] add_blinks (stochastic eye animation)
   → [WASM] Preset blending: expression channels (brows, eyes) blended with lip sync
diff --git a/examples/vanilla-avatar/README.md b/examples/vanilla-avatar/README.md
@@ -1,10 +1,10 @@
 # Vanilla Avatar (V2)
 
-Full 3D VRM avatar driven by AnimaSync V2 — the 52-dim student model engine. Lip sync, facial expressions, natural eye blinks, and body motion — all generated from a single audio stream via direct blendshape prediction.
+Full 3D VRM avatar driven by AnimaSync V2 — the 52-dim emotion model engine with 5-dim FiLM conditioning. Lip sync, facial expressions, natural eye blinks, and body motion — all generated from a single audio stream.
 
 ## What it demonstrates
 
-- **52-dim ARKit output**: Standard blendshape channels via student model direct prediction
+- **52-dim ARKit output**: Standard blendshape channels via emotion model with FiLM conditioning
 - **Lip sync**: Crisp mouth shapes with threshold-based sharpening
 - **Facial expressions**: Brows and eye area respond to vocal characteristics
 - **Eye animation**: Natural stochastic blinks injected by post-processing
diff --git a/examples/vanilla-avatar/index.html b/examples/vanilla-avatar/index.html
@@ -3,22 +3,22 @@
 <head>
   <meta charset="UTF-8">
   <meta name="viewport" content="width=device-width, initial-scale=1.0">
-  <title>AnimaSync — V2: Student Model Animation Data (52-dim)</title>
-  <meta name="description" content="AnimaSync V2 demo — student model lip sync engine with 52-dim ARKit blendshapes. Direct prediction, crisp mouth animation, real-time visualization.">
+  <title>AnimaSync — V2: Emotion Model Animation Data (52-dim)</title>
+  <meta name="description" content="AnimaSync V2 demo — emotion model lip sync engine with 52-dim ARKit blendshapes. 5-dim FiLM conditioning, crisp mouth animation, real-time visualization.">
   <meta name="robots" content="index, follow">
 
   <!-- Open Graph -->
   <meta property="og:site_name" content="AnimaSync">
-  <meta property="og:title" content="AnimaSync V2 Demo — Student Model Lip Sync">
-  <meta property="og:description" content="Upload audio and visualize 52 ARKit blendshapes from the V2 student model. Crisp mouth animation via direct prediction.">
+  <meta property="og:title" content="AnimaSync V2 Demo — Emotion Model Lip Sync">
+  <meta property="og:description" content="Upload audio and visualize 52 ARKit blendshapes from the V2 emotion model. 5-dim FiLM conditioning with crisp mouth animation.">
   <meta property="og:type" content="website">
   <meta property="og:url" content="https://animasync.quasar.ggls.dev/examples/vanilla-avatar/">
   <meta property="og:image" content="https://animasync.quasar.ggls.dev/assets/readme/hero-banner.svg">
 
   <!-- Twitter Card -->
   <meta name="twitter:card" content="summary_large_image">
-  <meta name="twitter:title" content="AnimaSync V2 Demo — Student Model Lip Sync">
-  <meta name="twitter:description" content="Upload audio and visualize 52 ARKit blendshapes from the V2 student model.">
+  <meta name="twitter:title" content="AnimaSync V2 Demo — Emotion Model Lip Sync">
+  <meta name="twitter:description" content="Upload audio and visualize 52 ARKit blendshapes from the V2 emotion model with FiLM conditioning.">
   <meta name="twitter:image" content="https://animasync.quasar.ggls.dev/assets/readme/hero-banner.svg">
 
   <!-- Canonical & Favicon -->
@@ -31,7 +31,7 @@
     "@context": "https://schema.org",
     "@type": "WebApplication",
     "name": "AnimaSync V2 Demo",
-    "description": "Student model lip sync engine demo — 52-dim ARKit blendshapes with direct prediction, running entirely in the browser via Rust/WASM.",
+    "description": "Emotion model lip sync engine demo — 52-dim ARKit blendshapes with 5-dim FiLM conditioning, running entirely in the browser via Rust/WASM.",
     "url": "https://animasync.quasar.ggls.dev/examples/vanilla-avatar/",
     "applicationCategory": "DeveloperApplication",
     "operatingSystem": "Browser",
@@ -183,7 +183,7 @@ <h2>Audio Input</h2>
 
   <!-- Right: Blendshapes -->
   <div class="card">
-    <h2>52 ARKit Blendshapes — V2 Student</h2>
+    <h2>52 ARKit Blendshapes — V2 Emotion</h2>
     <div class="bs-grid" id="bs-grid"></div>
   </div>
 </main>
@@ -195,7 +195,7 @@ <h2>52 ARKit Blendshapes — V2 Student</h2>
 
 <script type="module">
 // ================================================================
-// AnimaSync — V2 Student Model Example
+// AnimaSync — V2 Emotion Model Example
 // No 3D avatar, no Three.js. Pure audio → lip sync data (52-dim).
 // ================================================================
 
diff --git a/examples/vanilla-comparison/index.html b/examples/vanilla-comparison/index.html
@@ -30,7 +30,7 @@
     "@context": "https://schema.org",
     "@type": "WebApplication",
     "name": "AnimaSync V1 vs V2 Comparison",
-    "description": "Side-by-side comparison of V1 phoneme and V2 student model lip sync engines with dual 3D avatar rendering.",
+    "description": "Side-by-side comparison of V1 phoneme and V2 emotion model lip sync engines with dual 3D avatar rendering.",
     "url": "https://animasync.quasar.ggls.dev/examples/vanilla-comparison/",
     "applicationCategory": "DeveloperApplication",
     "operatingSystem": "Browser",
@@ -206,7 +206,7 @@ <h1>Anima<span>Sync</span></h1>
   </div>
   <div class="pane">
     <div class="pane-header">
-      <span class="pane-title v2">V2 — Student Model</span>
+      <span class="pane-title v2">V2 — Emotion Model</span>
       <span class="pane-dim">52-dim ARKit</span>
     </div>
     <div class="canvas-wrap">
diff --git a/index.html b/index.html
@@ -673,7 +673,7 @@ <h2 class="section-title">Install from npm</h2>
         <div class="pkg-name">@goodganglabs/lipsync-wasm-v2</div>
         <span class="pkg-tag pkg-tag-full">Lightweight</span>
       </div>
-      <p class="pkg-desc">Student distillation model &mdash; direct 52-dim ARKit blendshape prediction with 5-dim emotion control (FiLM conditioning).</p>
+      <p class="pkg-desc">Emotion model &mdash; 52-dim ARKit blendshape prediction with 5-dim FiLM conditioning (neutral, joy, anger, sadness, surprise).</p>
       <div class="pkg-meta">
         <div class="pkg-meta-item">Output: <span>52-dim</span> ARKit</div>
         <div class="pkg-meta-item">Emotion: <span>5-dim</span> FiLM conditioning</div>
@@ -714,8 +714,8 @@ <h3>Phoneme Visualization</h3>
     <a href="examples/vanilla-avatar/" class="example-card">
       <div class="example-card-body">
         <div class="card-badge">V2 Engine</div>
-        <h3>Student Model Demo</h3>
-        <p>V2 student distillation model &mdash; 52 ARKit blendshapes with direct prediction. Crisp mouth, real-time rendering.</p>
+        <h3>Emotion Model Demo</h3>
+        <p>V2 emotion model &mdash; 52 ARKit blendshapes with 5-dim FiLM conditioning. Emotion-aware lip sync, real-time rendering.</p>
         <span class="card-link">Try it &#8594;</span>
       </div>
     </a>
@@ -747,7 +747,7 @@ <h2 class="section-title">Choose your engine</h2>
     </thead>
     <tbody>
       <tr><td>Output</td><td>111-dim ARKit blendshapes</td><td>52-dim ARKit blendshapes</td></tr>
-      <tr><td>Architecture</td><td>Phoneme classification + viseme mapping</td><td>Student distillation (direct)</td></tr>
+      <tr><td>Architecture</td><td>Phoneme classification + viseme mapping</td><td>Emotion model + FiLM conditioning</td></tr>
       <tr><td>Post-processing</td><td>OneEuroFilter + anatomical constraints</td><td>crisp_mouth + fade + auto-blink</td></tr>
       <tr><td>Idle expressions</td><td>Built-in IdleExpressionGenerator</td><td>Blink injection in post-process</td></tr>
       <tr><td>Voice activity</td><td>Built-in VoiceActivityDetector</td><td>&mdash;</td></tr>
diff --git a/llms-full.txt b/llms-full.txt
@@ -153,7 +153,7 @@ interface ProcessResult {
 |---------|-------------------|-------------------|
 | npm package | @goodganglabs/lipsync-wasm-v1 | @goodganglabs/lipsync-wasm-v2 |
 | Output dimension | 111-dim ARKit blendshapes | 52-dim ARKit blendshapes |
-| Model architecture | Phoneme classification -> viseme mapping | Student distillation (direct prediction) |
+| Model architecture | Phoneme classification -> viseme mapping | Emotion model + FiLM conditioning |
 | Post-processing | OneEuroFilter + anatomical constraints | crisp_mouth + fade + auto-blink |
 | Expression generation | Built-in IdleExpressionGenerator | Blink injection in post-process |
 | VRM mode | getVrmFrame() + convert_arkit_to_vrm() for VRM 18-dim | getFrame() only (52-dim ARKit) |
@@ -172,7 +172,7 @@ interface ProcessResult {
 ```
 Audio 16kHz PCM
   -> [WASM] librosa-compatible features: 141-dim @30fps
-  -> [JS]   ONNX student model -> 52-dim (lip sync + expressions)
+  -> [JS]   ONNX emotion model + FiLM conditioning -> 52-dim (lip sync + expressions)
   -> [WASM] crisp_mouth (mouth sharpening) -> fade_in_out (natural onset/offset)
   -> [WASM] add_blinks (stochastic eye animation)
   -> [WASM] Preset blending: expression channels blended with lip sync
@@ -214,7 +214,7 @@ Tongue: tongueOut
 |---------|-------------|-----|
 | Step-by-Step Guide | 6-step interactive tutorial: VRM + AnimaSync V1 lip sync with live demos (CDN 0.4.5, VRM mode auto-detect, idle eye blink, audio-synced playback, LoopPingPong idle, asymmetric crossfade) | https://animasync.quasar.ggls.dev/examples/guide/ |
 | V1 Data | V1 phoneme engine — 52 ARKit blendshapes visualization | https://animasync.quasar.ggls.dev/examples/vanilla-basic/ |
-| V2 Data | V2 student model — 52 ARKit direct prediction | https://animasync.quasar.ggls.dev/examples/vanilla-avatar/ |
+| V2 Data | V2 emotion model — 52 ARKit with 5-dim FiLM conditioning | https://animasync.quasar.ggls.dev/examples/vanilla-avatar/ |
 | V1 vs V2 | Side-by-side dual avatar comparison | https://animasync.quasar.ggls.dev/examples/vanilla-comparison/ |
 
 Run locally:
diff --git a/llms.txt b/llms.txt
@@ -14,7 +14,7 @@ AnimaSync extracts emotion from speech and generates lip sync, facial expression
 ## Two Engine Versions
 
 - **V1 (Recommended)**: Phoneme classification, 111-dim ARKit output, built-in VAD
-- **V2 (Lightweight)**: Student distillation model, 52-dim ARKit output, direct prediction, 5-dim emotion control (FiLM conditioning)
+- **V2 (Emotion)**: Emotion model, 52-dim ARKit output, 5-dim FiLM conditioning (neutral, joy, anger, sadness, surprise)
 
 ## Quick Start
 

Original file line number	Diff line number	Diff line change
`@@ -53,7 +53,7 @@`
`53`	`53`	`},`
`54`	`54`	`{`
`55`	`55`	`"name": "@goodganglabs/lipsync-wasm-v2",`
`56`		`- "description": "V2 engine — 52-dim ARKit blendshapes via student distillation with 5-dim emotion control (FiLM conditioning).",`
	`56`	`+ "description": "V2 engine — 52-dim ARKit blendshapes via emotion model with 5-dim FiLM conditioning (neutral, joy, anger, sadness, surprise).",`
`57`	`57`	`"url": "https://www.npmjs.com/package/@goodganglabs/lipsync-wasm-v2"`
`58`	`58`	`}`
`59`	`59`	`]`