Spaces:

ConvxO2
/

Who-Spoke-When

Running

App Files Files Community

Who-Spoke-When / static /index.html

ConvxO2

Fix Spaces UI API endpoint default to same-origin

6aa584f 7 days ago

raw

history blame contribute delete

17.3 kB

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8" />
	<meta name="viewport" content="width=device-width, initial-scale=1.0"/>
	<title>Speaker Diarization System</title>
	<style>
	@import url('https://fonts.googleapis.com/css2?family=JetBrains+Mono:wght@300;400;500;700&family=Space+Grotesk:wght@300;400;600;700&display=swap');

	:root {
	--bg: #090c10;
	--surface: #0f1318;
	--surface2: #151b23;
	--border: #1e2730;
	--accent: #00d4ff;
	--accent2: #7c3aed;
	--green: #22d3a0;
	--yellow: #f59e0b;
	--red: #ef4444;
	--text: #e2e8f0;
	--muted: #64748b;
	--font-mono: 'JetBrains Mono', monospace;
	--font-sans: 'Space Grotesk', sans-serif;
	}

	* { margin: 0; padding: 0; box-sizing: border-box; }

	body {
	font-family: var(--font-sans);
	background: var(--bg);
	color: var(--text);
	min-height: 100vh;
	overflow-x: hidden;
	}

	/* Grid bg */
	body::before {
	content: '';
	position: fixed;
	inset: 0;
	background-image:
	linear-gradient(rgba(0, 212, 255, 0.03) 1px, transparent 1px),
	linear-gradient(90deg, rgba(0, 212, 255, 0.03) 1px, transparent 1px);
	background-size: 40px 40px;
	pointer-events: none;
	z-index: 0;
	}

	.container {
	position: relative;
	z-index: 1;
	max-width: 1100px;
	margin: 0 auto;
	padding: 2rem 1.5rem;
	}

	header {
	text-align: center;
	margin-bottom: 3rem;
	}

	.badge {
	display: inline-block;
	background: rgba(0, 212, 255, 0.1);
	border: 1px solid rgba(0, 212, 255, 0.3);
	color: var(--accent);
	font-family: var(--font-mono);
	font-size: 0.72rem;
	letter-spacing: 0.15em;
	padding: 4px 12px;
	border-radius: 100px;
	margin-bottom: 1rem;
	}

	h1 {
	font-size: clamp(2rem, 5vw, 3.2rem);
	font-weight: 700;
	letter-spacing: -0.02em;
	background: linear-gradient(135deg, #fff 30%, var(--accent));
	-webkit-background-clip: text;
	-webkit-text-fill-color: transparent;
	line-height: 1.15;
	}

	.subtitle {
	color: var(--muted);
	font-size: 1rem;
	margin-top: 0.75rem;
	font-weight: 300;
	}

	/* Cards */
	.card {
	background: var(--surface);
	border: 1px solid var(--border);
	border-radius: 12px;
	padding: 1.5rem;
	margin-bottom: 1.5rem;
	}

	.card-title {
	font-size: 0.8rem;
	font-family: var(--font-mono);
	letter-spacing: 0.12em;
	color: var(--accent);
	text-transform: uppercase;
	margin-bottom: 1.2rem;
	display: flex;
	align-items: center;
	gap: 8px;
	}

	.card-title::before {
	content: '▸';
	font-size: 0.9rem;
	}

	/* Upload zone */
	.upload-zone {
	border: 2px dashed var(--border);
	border-radius: 10px;
	padding: 2.5rem;
	text-align: center;
	cursor: pointer;
	transition: all 0.25s;
	position: relative;
	}

	.upload-zone:hover, .upload-zone.drag-over {
	border-color: var(--accent);
	background: rgba(0, 212, 255, 0.04);
	}

	.upload-zone input[type="file"] {
	position: absolute;
	inset: 0;
	opacity: 0;
	cursor: pointer;
	}

	.upload-icon {
	font-size: 2.5rem;
	margin-bottom: 0.75rem;
	opacity: 0.6;
	}

	.upload-text {
	color: var(--muted);
	font-size: 0.9rem;
	}

	.upload-text strong {
	color: var(--accent);
	}

	/* Controls */
	.controls {
	display: grid;
	grid-template-columns: 1fr 1fr auto;
	gap: 1rem;
	margin-top: 1rem;
	align-items: end;
	}

	.field label {
	display: block;
	font-size: 0.75rem;
	font-family: var(--font-mono);
	color: var(--muted);
	margin-bottom: 6px;
	letter-spacing: 0.08em;
	}

	.field input, .field select {
	width: 100%;
	background: var(--surface2);
	border: 1px solid var(--border);
	color: var(--text);
	font-family: var(--font-mono);
	font-size: 0.9rem;
	padding: 10px 12px;
	border-radius: 8px;
	outline: none;
	transition: border-color 0.2s;
	}

	.field input:focus, .field select:focus {
	border-color: var(--accent);
	}

	.btn-primary {
	background: var(--accent);
	color: #000;
	font-family: var(--font-sans);
	font-weight: 700;
	font-size: 0.9rem;
	border: none;
	padding: 10px 24px;
	border-radius: 8px;
	cursor: pointer;
	transition: all 0.2s;
	white-space: nowrap;
	}

	.btn-primary:hover { filter: brightness(1.1); transform: translateY(-1px); }
	.btn-primary:disabled { opacity: 0.4; cursor: not-allowed; transform: none; }

	/* Progress */
	.progress-bar {
	height: 4px;
	background: var(--border);
	border-radius: 99px;
	overflow: hidden;
	margin-top: 1rem;
	display: none;
	}

	.progress-fill {
	height: 100%;
	background: linear-gradient(90deg, var(--accent), var(--accent2));
	width: 0%;
	transition: width 0.4s;
	animation: progress-pulse 1.5s ease-in-out infinite;
	}

	@keyframes progress-pulse {
	0%, 100% { opacity: 1; }
	50% { opacity: 0.6; }
	}

	/* Stats row */
	.stats-row {
	display: grid;
	grid-template-columns: repeat(4, 1fr);
	gap: 1rem;
	margin-bottom: 1.5rem;
	}

	.stat {
	background: var(--surface);
	border: 1px solid var(--border);
	border-radius: 10px;
	padding: 1rem 1.2rem;
	}

	.stat-val {
	font-family: var(--font-mono);
	font-size: 1.8rem;
	font-weight: 700;
	color: var(--accent);
	}

	.stat-label {
	font-size: 0.73rem;
	color: var(--muted);
	margin-top: 4px;
	letter-spacing: 0.06em;
	}

	/* Timeline */
	#timeline-container {
	margin-bottom: 1rem;
	}

	.timeline-ruler {
	display: flex;
	justify-content: space-between;
	font-family: var(--font-mono);
	font-size: 0.68rem;
	color: var(--muted);
	margin-bottom: 6px;
	padding: 0 2px;
	}

	.timeline-track {
	height: 48px;
	background: var(--surface2);
	border-radius: 8px;
	position: relative;
	overflow: hidden;
	border: 1px solid var(--border);
	margin-bottom: 8px;
	}

	.track-label {
	font-family: var(--font-mono);
	font-size: 0.68rem;
	color: var(--muted);
	position: absolute;
	left: 8px;
	top: 50%;
	transform: translateY(-50%);
	z-index: 2;
	text-shadow: 0 0 8px var(--bg);
	}

	.timeline-segment {
	position: absolute;
	height: 100%;
	border-radius: 4px;
	opacity: 0.9;
	cursor: pointer;
	transition: opacity 0.15s, filter 0.15s;
	display: flex;
	align-items: center;
	justify-content: center;
	font-family: var(--font-mono);
	font-size: 0.65rem;
	color: rgba(0,0,0,0.85);
	font-weight: 700;
	overflow: hidden;
	white-space: nowrap;
	}

	.timeline-segment:hover {
	opacity: 1;
	filter: brightness(1.15);
	z-index: 5;
	}

	/* Segment table */
	.seg-table {
	width: 100%;
	border-collapse: collapse;
	font-family: var(--font-mono);
	font-size: 0.82rem;
	}

	.seg-table th {
	text-align: left;
	padding: 8px 12px;
	font-size: 0.7rem;
	letter-spacing: 0.1em;
	color: var(--muted);
	border-bottom: 1px solid var(--border);
	}

	.seg-table td {
	padding: 9px 12px;
	border-bottom: 1px solid rgba(255,255,255,0.04);
	vertical-align: middle;
	}

	.seg-table tr:last-child td { border-bottom: none; }
	.seg-table tr:hover td { background: rgba(255,255,255,0.02); }

	.speaker-dot {
	display: inline-block;
	width: 8px;
	height: 8px;
	border-radius: 50%;
	margin-right: 8px;
	}

	/* Log */
	#log {
	font-family: var(--font-mono);
	font-size: 0.78rem;
	color: var(--muted);
	background: var(--surface2);
	border-radius: 8px;
	padding: 1rem;
	max-height: 160px;
	overflow-y: auto;
	line-height: 1.7;
	}

	.log-info { color: var(--accent); }
	.log-success{ color: var(--green); }
	.log-error { color: var(--red); }
	.log-warn { color: var(--yellow); }

	.hidden { display: none !important; }

	@media (max-width: 640px) {
	.controls { grid-template-columns: 1fr; }
	.stats-row { grid-template-columns: 1fr 1fr; }
	}
	</style>
	</head>
	<body>
	<div class="container">

	<header>
	<div class="badge">ECAPA-TDNN + AHC · FASTAPI</div>
	<h1>Speaker Diarization System</h1>
	<p class="subtitle">Who spoke when — multi-speaker audio segmentation & labeling</p>
	</header>

	<!-- Upload Card -->
	<div class="card">
	<div class="card-title">Audio Input</div>
	<div class="upload-zone" id="dropzone">
	<input type="file" id="audioFile" accept=".wav,.mp3,.flac,.ogg,.m4a,.webm" />
	<div class="upload-icon">🎙</div>
	<div class="upload-text"><strong>Drop audio file</strong> or click to browse</div>
	<div class="upload-text" style="margin-top:4px;font-size:0.78rem;" id="filename-display">WAV · MP3 · FLAC · OGG · M4A</div>
	</div>

	<div class="controls">
	<div class="field">
	<label>API ENDPOINT</label>
	<input type="text" id="apiUrl" value="/diarize" />
	</div>
	<div class="field">
	<label>SPEAKERS (blank = auto)</label>
	<input type="number" id="numSpeakers" min="1" max="20" placeholder="auto-detect" />
	</div>
	<button class="btn-primary" id="runBtn" onclick="runDiarization()" disabled>
	▶ Run
	</button>
	</div>

	<div class="progress-bar" id="progressBar">
	<div class="progress-fill" id="progressFill"></div>
	</div>
	</div>

	<!-- Results (hidden until run) -->
	<div id="results" class="hidden">

	<div class="stats-row" id="statsRow"></div>

	<!-- Timeline -->
	<div class="card">
	<div class="card-title">Speaker Timeline</div>
	<div class="timeline-ruler" id="timelineRuler"></div>
	<div id="timelineTracks"></div>
	</div>

	<!-- Segment Table -->
	<div class="card">
	<div class="card-title">Segments</div>
	<div style="overflow-x:auto;">
	<table class="seg-table">
	<thead>
	<tr>
	<th>#</th><th>SPEAKER</th><th>START</th><th>END</th><th>DURATION</th>
	</tr>
	</thead>
	<tbody id="segTableBody"></tbody>
	</table>
	</div>
	</div>

	</div>

	<!-- Log -->
	<div class="card">
	<div class="card-title">Log</div>
	<div id="log"><span class="log-info">// Ready. Upload an audio file to begin.</span></div>
	</div>

	</div>

	<script>
	const SPEAKER_COLORS = [
	'#00d4ff','#7c3aed','#22d3a0','#f59e0b',
	'#ec4899','#3b82f6','#84cc16','#f97316',
	'#06b6d4','#a855f7',
	];

	let selectedFile = null;

	// Default to same-origin API on hosted deployments (e.g., Hugging Face Spaces).
	const apiUrlInput = document.getElementById('apiUrl');
	if (apiUrlInput && (!apiUrlInput.value \|\| apiUrlInput.value.includes('localhost'))) {
	apiUrlInput.value = `/diarize`;
	}

	// ── File Handling ──────────────────────────────────────────────────────────
	document.getElementById('audioFile').addEventListener('change', (e) => {
	const file = e.target.files[0];
	if (!file) return;
	selectedFile = file;
	document.getElementById('filename-display').textContent = `📁 ${file.name} (${(file.size/1024/1024).toFixed(2)} MB)`;
	document.getElementById('runBtn').disabled = false;
	log(`File selected: ${file.name}`, 'info');
	});

	// Drag & Drop
	const dz = document.getElementById('dropzone');
	dz.addEventListener('dragover', e => { e.preventDefault(); dz.classList.add('drag-over'); });
	dz.addEventListener('dragleave', () => dz.classList.remove('drag-over'));
	dz.addEventListener('drop', e => {
	e.preventDefault();
	dz.classList.remove('drag-over');
	const file = e.dataTransfer.files[0];
	if (file) {
	document.getElementById('audioFile').files = e.dataTransfer.files;
	document.getElementById('audioFile').dispatchEvent(new Event('change'));
	}
	});

	// ── Log ────────────────────────────────────────────────────────────────────
	function log(msg, type = '') {
	const el = document.getElementById('log');
	const cls = type ? `log-${type}` : '';
	const ts = new Date().toLocaleTimeString('en', { hour12: false });
	el.innerHTML += `<br><span class="${cls}">[${ts}] ${msg}</span>`;
	el.scrollTop = el.scrollHeight;
	}

	// ── Run Diarization ────────────────────────────────────────────────────────
	async function runDiarization() {
	if (!selectedFile) return;

	const btn = document.getElementById('runBtn');
	const pb = document.getElementById('progressBar');
	const pf = document.getElementById('progressFill');

	btn.disabled = true;
	pb.style.display = 'block';
	pf.style.width = '20%';
	document.getElementById('results').classList.add('hidden');
	log('Uploading audio and running diarization...', 'info');

	const formData = new FormData();
	formData.append('file', selectedFile);
	const ns = document.getElementById('numSpeakers').value;
	if (ns) formData.append('num_speakers', ns);

	const url = document.getElementById('apiUrl').value;

	try {
	pf.style.width = '50%';
	const resp = await fetch(url, { method: 'POST', body: formData });

	pf.style.width = '90%';

	if (!resp.ok) {
	const err = await resp.json().catch(() => ({ detail: resp.statusText }));
	throw new Error(err.detail \|\| `HTTP ${resp.status}`);
	}

	const data = await resp.json();
	pf.style.width = '100%';

	log(`Done — ${data.num_speakers} speaker(s), ${data.segments.length} segments, ${data.processing_time.toFixed(2)}s`, 'success');
	renderResults(data);

	} catch (e) {
	log(`Error: ${e.message}`, 'error');
	} finally {
	setTimeout(() => { pb.style.display = 'none'; pf.style.width = '0%'; }, 800);
	btn.disabled = false;
	}
	}

	// ── Render Results ─────────────────────────────────────────────────────────
	function renderResults(data) {
	document.getElementById('results').classList.remove('hidden');

	// Stats
	const stats = [
	{ val: data.num_speakers, label: 'SPEAKERS' },
	{ val: data.segments.length, label: 'SEGMENTS' },
	{ val: data.audio_duration.toFixed(1) + 's', label: 'DURATION' },
	{ val: data.processing_time.toFixed(2) + 's', label: 'PROC TIME' },
	];
	document.getElementById('statsRow').innerHTML = stats.map(s =>
	`<div class="stat">
	<div class="stat-val">${s.val}</div>
	<div class="stat-label">${s.label}</div>
	</div>`
	).join('');

	// Build speaker→color map
	const colorMap = {};
	data.speakers.forEach((sp, i) => {
	colorMap[sp] = SPEAKER_COLORS[i % SPEAKER_COLORS.length];
	});

	// Timeline
	const duration = data.audio_duration;
	const ruler = document.getElementById('timelineRuler');
	const ticks = 8;
	ruler.innerHTML = Array.from({ length: ticks + 1 }, (_, i) =>
	`<span>${fmtTime(duration * i / ticks)}</span>`
	).join('');

	// One track per speaker
	const tracksEl = document.getElementById('timelineTracks');
	tracksEl.innerHTML = '';

	data.speakers.forEach(sp => {
	const track = document.createElement('div');
	track.className = 'timeline-track';
	track.innerHTML = `<span class="track-label">${sp}</span>`;

	const spSegs = data.segments.filter(s => s.speaker === sp);
	spSegs.forEach(seg => {
	const left = (seg.start / duration) * 100;
	const width = (seg.duration / duration) * 100;
	const seg_el = document.createElement('div');
	seg_el.className = 'timeline-segment';
	seg_el.style.cssText = `left:${left}%;width:${Math.max(width, 0.3)}%;background:${colorMap[sp]};`;
	seg_el.title = `${sp}: ${seg.start.toFixed(2)}s – ${seg.end.toFixed(2)}s`;
	if (width > 3) seg_el.textContent = fmtTime(seg.duration);
	track.appendChild(seg_el);
	});

	tracksEl.appendChild(track);
	});

	// Segment table
	const tbody = document.getElementById('segTableBody');
	tbody.innerHTML = data.segments.map((seg, i) =>
	`<tr>
	<td style="color:var(--muted)">${i + 1}</td>
	<td>
	<span class="speaker-dot" style="background:${colorMap[seg.speaker]}"></span>
	${seg.speaker}
	</td>
	<td>${seg.start.toFixed(3)}</td>
	<td>${seg.end.toFixed(3)}</td>
	<td>${seg.duration.toFixed(3)}s</td>
	</tr>`
	).join('');
	}

	function fmtTime(sec) {
	const m = Math.floor(sec / 60);
	const s = (sec % 60).toFixed(1).padStart(4, '0');
	return `${m}:${s}`;
	}
	</script>
	</body>
	</html>