ykhrustalev's picture
Fix cross-origin isolation to resolve TTS crash and slow ASR on HuggingFace (#4)
04fa39a
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>LFM2.5-Audio - WebGPU Demo</title>
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@300;400;500;600;700&display=swap" rel="stylesheet">
<style>
* {
box-sizing: border-box;
margin: 0;
padding: 0;
}
body {
font-family: 'JetBrains Mono', monospace;
background: #fafafa;
color: #0a0a0a;
min-height: 100vh;
display: flex;
flex-direction: column;
-webkit-font-smoothing: antialiased;
-moz-osx-font-smoothing: grayscale;
}
header {
background: #fff;
padding: 1.25rem 1.5rem;
border-bottom: 1px solid rgba(0, 0, 0, 0.08);
display: flex;
align-items: center;
gap: 1rem;
}
header .logo {
display: flex;
align-items: center;
gap: 0.75rem;
}
header .logo svg {
width: 28px;
height: 28px;
}
header h1 {
font-size: 1.1rem;
font-weight: 600;
letter-spacing: -0.02em;
}
header .badge {
font-size: 0.65rem;
padding: 0.2rem 0.5rem;
border-radius: 4px;
font-weight: 500;
background: #0a0a0a;
color: #fff;
letter-spacing: 0.02em;
}
.controls {
display: flex;
gap: 1rem;
padding: 1rem 1.5rem;
background: #fff;
border-bottom: 1px solid rgba(0, 0, 0, 0.08);
flex-wrap: wrap;
align-items: center;
}
.controls label {
display: flex;
align-items: center;
gap: 0.5rem;
font-size: 0.8rem;
font-weight: 500;
color: #666;
}
.controls select {
padding: 0.5rem 0.75rem;
border-radius: 6px;
border: 1px solid rgba(0, 0, 0, 0.12);
background: #fff;
color: #0a0a0a;
font-size: 0.8rem;
font-family: 'JetBrains Mono', monospace;
cursor: pointer;
transition: all 0.2s ease;
}
.controls select:hover {
border-color: rgba(0, 0, 0, 0.24);
}
.controls select:focus {
outline: none;
border-color: #0a0a0a;
}
.controls button {
padding: 0.5rem 1rem;
border-radius: 6px;
border: 1px solid #0a0a0a;
background: #0a0a0a;
color: #fff;
font-size: 0.8rem;
font-family: 'JetBrains Mono', monospace;
font-weight: 500;
cursor: pointer;
transition: all 0.2s ease;
}
.controls button:disabled {
opacity: 0.4;
cursor: not-allowed;
}
.controls button:hover:not(:disabled) {
background: #333;
}
.controls button.secondary {
background: transparent;
color: #0a0a0a;
}
.controls button.secondary:hover:not(:disabled) {
background: rgba(0, 0, 0, 0.05);
}
.storage-info {
display: flex;
align-items: center;
gap: 0.5rem;
margin-left: auto;
font-size: 0.75rem;
color: #888;
}
.text-btn {
background: none;
border: none;
padding: 0.25rem 0.5rem;
font-size: 0.7rem;
font-family: 'JetBrains Mono', monospace;
cursor: pointer;
color: #888;
transition: color 0.2s ease;
}
.text-btn:hover:not(:disabled) {
color: #0a0a0a;
}
.text-btn.danger {
color: #dc2626;
}
.text-btn.danger:hover:not(:disabled) {
color: #b91c1c;
}
.text-btn:disabled {
color: #ccc;
cursor: not-allowed;
}
#status {
padding: 0.6rem 1.5rem;
background: #f5f5f5;
font-size: 0.75rem;
color: #666;
border-bottom: 1px solid rgba(0, 0, 0, 0.05);
}
#status.error {
background: #fef2f2;
color: #dc2626;
}
#status.success {
background: #f0fdf4;
color: #16a34a;
}
.chat-container {
flex: 1;
overflow-y: auto;
padding: 1.5rem;
display: flex;
flex-direction: column;
gap: 1rem;
background: #fafafa;
}
.message {
max-width: 75%;
padding: 1rem 1.25rem;
border-radius: 12px;
line-height: 1.6;
white-space: pre-wrap;
font-size: 0.85rem;
}
.message.user {
align-self: flex-end;
background: #0a0a0a;
color: #fff;
}
.message.assistant {
align-self: flex-start;
background: #fff;
border: 1px solid rgba(0, 0, 0, 0.08);
}
.message.generating {
opacity: 0.7;
}
.message audio {
margin-top: 0.75rem;
width: 100%;
max-width: 360px;
min-height: 40px;
height: auto;
border-radius: 8px;
display: block;
}
.input-area {
padding: 1.25rem 1.5rem;
background: #fff;
border-top: 1px solid rgba(0, 0, 0, 0.08);
}
.audio-preview {
display: flex;
gap: 0.5rem;
margin-bottom: 0.75rem;
flex-wrap: wrap;
}
.audio-preview-item {
position: relative;
display: flex;
align-items: center;
gap: 0.5rem;
padding: 0.5rem 0.75rem;
background: #f5f5f5;
border-radius: 8px;
border: 1px solid rgba(0, 0, 0, 0.08);
font-size: 0.8rem;
}
.audio-preview-item .audio-icon {
font-size: 1rem;
}
.audio-preview-item .audio-name {
max-width: 140px;
overflow: hidden;
text-overflow: ellipsis;
white-space: nowrap;
font-weight: 500;
}
.audio-preview-item .audio-info {
font-size: 0.7rem;
color: #888;
}
.audio-preview-item .remove-btn {
position: absolute;
top: -6px;
right: -6px;
width: 18px;
height: 18px;
border-radius: 50%;
background: #0a0a0a;
color: white;
border: none;
cursor: pointer;
font-size: 10px;
display: flex;
align-items: center;
justify-content: center;
transition: background 0.2s ease;
}
.audio-preview-item .remove-btn:hover {
background: #333;
}
.input-row {
display: flex;
gap: 0.5rem;
}
#userInput {
flex: 1;
padding: 0.75rem 1rem;
border-radius: 8px;
border: 1px solid rgba(0, 0, 0, 0.12);
background: #fff;
color: #0a0a0a;
font-size: 0.85rem;
font-family: 'JetBrains Mono', monospace;
resize: none;
transition: border-color 0.2s ease;
}
#userInput:focus {
outline: none;
border-color: #0a0a0a;
}
#userInput::placeholder {
color: #aaa;
}
.input-buttons {
display: flex;
gap: 0.35rem;
}
#audioBtn, #recordBtn {
padding: 0.6rem;
border-radius: 8px;
border: 1px solid rgba(0, 0, 0, 0.12);
background: #fff;
color: #0a0a0a;
cursor: pointer;
font-size: 1.1rem;
transition: all 0.2s ease;
display: flex;
align-items: center;
justify-content: center;
width: 42px;
height: 42px;
}
#audioBtn:hover:not(:disabled), #recordBtn:hover:not(:disabled) {
background: #f5f5f5;
border-color: rgba(0, 0, 0, 0.24);
}
#audioBtn:disabled, #recordBtn:disabled {
opacity: 0.4;
cursor: not-allowed;
}
#recordBtn.recording {
background: #dc2626;
border-color: #dc2626;
color: #fff;
animation: pulse 1.5s infinite;
}
@keyframes pulse {
0%, 100% { opacity: 1; transform: scale(1); }
50% { opacity: 0.8; transform: scale(0.98); }
}
#sendBtn {
padding: 0.6rem 1.25rem;
border-radius: 8px;
border: none;
background: #0a0a0a;
color: white;
font-weight: 500;
font-family: 'JetBrains Mono', monospace;
font-size: 0.8rem;
cursor: pointer;
transition: all 0.2s ease;
}
#sendBtn:hover:not(:disabled) {
background: #333;
}
#sendBtn:disabled {
opacity: 0.4;
cursor: not-allowed;
}
.stats {
font-size: 0.65rem;
color: #888;
margin-top: 0.5rem;
font-weight: 400;
}
.progress-bar {
position: relative;
height: 3px;
background: #eee;
overflow: hidden;
}
.progress-fill {
height: 100%;
background: #0a0a0a;
width: 0%;
transition: width 0.3s ease;
}
.progress-text {
position: absolute;
top: 8px;
left: 1.5rem;
font-size: 0.7rem;
color: #666;
}
#audioInput {
display: none;
}
.drop-overlay {
display: none;
position: fixed;
inset: 0;
background: rgba(255, 255, 255, 0.95);
border: 3px dashed #0a0a0a;
z-index: 1000;
align-items: center;
justify-content: center;
font-size: 1.25rem;
color: #0a0a0a;
font-weight: 500;
}
.drop-overlay.active {
display: flex;
}
.info-box {
background: #fff;
border: 1px solid rgba(0, 0, 0, 0.08);
border-radius: 12px;
padding: 1.5rem;
font-size: 0.8rem;
line-height: 1.7;
}
.info-box h3 {
margin-bottom: 0.75rem;
font-size: 0.9rem;
font-weight: 600;
letter-spacing: -0.01em;
}
.info-box p {
color: #444;
margin-bottom: 1rem;
}
.info-box ul {
margin-left: 1.25rem;
color: #444;
}
.info-box li {
margin-bottom: 0.5rem;
}
.info-box strong {
color: #0a0a0a;
font-weight: 600;
}
.info-box .note {
margin-top: 1rem;
padding-top: 1rem;
border-top: 1px solid rgba(0, 0, 0, 0.06);
font-size: 0.75rem;
color: #888;
}
/* Spinner */
.spinner {
display: none;
align-items: center;
gap: 0.5rem;
padding: 0.75rem 1rem;
background: #fff;
border: 1px solid rgba(0, 0, 0, 0.08);
border-radius: 8px;
font-size: 0.8rem;
color: #666;
}
.spinner.active {
display: flex;
}
.spinner-icon {
width: 16px;
height: 16px;
border: 2px solid rgba(0, 0, 0, 0.1);
border-top-color: #0a0a0a;
border-radius: 50%;
animation: spin 0.8s linear infinite;
}
@keyframes spin {
to { transform: rotate(360deg); }
}
.spinner-text {
flex: 1;
}
.spinner-stats {
font-size: 0.7rem;
color: #888;
font-variant-numeric: tabular-nums;
}
/* Scrollbar styling */
::-webkit-scrollbar {
width: 6px;
}
::-webkit-scrollbar-track {
background: transparent;
}
::-webkit-scrollbar-thumb {
background: rgba(0, 0, 0, 0.15);
border-radius: 3px;
}
::-webkit-scrollbar-thumb:hover {
background: rgba(0, 0, 0, 0.25);
}
</style>
</head>
<body>
<header>
<div class="logo">
<svg viewBox="0 0 100 100" fill="none" xmlns="http://www.w3.org/2000/svg">
<circle cx="50" cy="50" r="45" stroke="#0a0a0a" stroke-width="6"/>
<circle cx="50" cy="50" r="20" fill="#0a0a0a"/>
</svg>
<h1>LFM2.5-Audio <span class="badge">WebGPU</span></h1>
</div>
</header>
<div class="controls">
<label>
Model
<select id="modelSelect"></select>
</label>
<button id="loadBtn">Load Model</button>
<label>
Mode
<select id="audioModeSelect" title="Audio interaction mode">
<option value="asr">ASR (Speech → Text)</option>
<option value="tts">TTS (Text → Speech)</option>
<option value="interleaved" selected>Interleaved</option>
</select>
</label>
<button id="clearBtn" class="secondary" title="Reset conversation state (clears KV cache)">New Chat</button>
<div class="storage-info">
<span id="cacheInfo"></span>
<button id="clearCacheBtn" class="text-btn danger" title="Delete downloaded model files">Clear Cache</button>
</div>
</div>
<div id="progressBar" class="progress-bar" style="display: none;">
<div id="progressFill" class="progress-fill"></div>
<span id="progressText" class="progress-text"></span>
</div>
<div id="status">Select a model and click "Load Model" to start</div>
<div class="chat-container" id="chatContainer">
<div class="info-box">
<h3>LFM2.5-Audio Demo</h3>
<p>Speech recognition, text-to-speech, and conversational audio running entirely in your browser.</p>
<ul>
<li><strong>ASR</strong> — Record or upload audio to transcribe to text</li>
<li><strong>TTS</strong> — Type text to generate natural speech</li>
<li><strong>Interleaved</strong> — Have a spoken conversation with the model (multi-turn with stateful KV cache)</li>
</ul>
<p class="note">Requires WebGPU-enabled browser (Chrome/Edge 113+). Enable at chrome://flags if needed. Use "New Chat" to reset conversation.</p>
</div>
</div>
<div class="input-area">
<div class="spinner" id="spinner">
<div class="spinner-icon"></div>
<span class="spinner-text" id="spinnerText">Processing...</span>
<span class="spinner-stats" id="spinnerStats"></span>
</div>
<div class="audio-preview" id="audioPreview"></div>
<div class="input-row">
<textarea id="userInput" rows="2" placeholder="Type a message or record audio..." disabled></textarea>
<div class="input-buttons">
<button id="audioBtn" title="Upload audio file" disabled>📁</button>
<button id="recordBtn" title="Record from microphone" disabled>🎤</button>
<button id="sendBtn" disabled>Send</button>
</div>
</div>
<input type="file" id="audioInput" accept="audio/*">
</div>
<div class="drop-overlay" id="dropOverlay">Drop audio file here</div>
<script src="./coi-serviceworker.js"></script>
<script type="module" src="./main.js"></script>
</body>
</html>