ASR-demo/realtime_asr_optimization_demo/static/index.html

120 lines
6.4 KiB
HTML
Raw Normal View History

2026-09-10 05:47:09 +00:00
<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>语音识别体验 Demo</title>
<link rel="stylesheet" href="style.css?v=210">
</head>
<body>
<div class="layout">
<!-- 左侧面板 -->
<aside class="panel-left">
<header>
<h1>🎤 语音识别体验</h1>
<p class="subtitle">实时语音识别 · 句子模式</p>
</header>
<section class="card">
<h2>识别配置</h2>
<div class="form-stack">
<div class="form-group">
<label for="engineModel">引擎模型</label>
<input type="text" id="engineModel" value="Qwen/Qwen3-ASR-0.6B">
</div>
<div class="form-group">
<label for="modelServiceUrl">vLLM 服务地址</label>
<input type="text" id="modelServiceUrl" value="http://127.0.0.1:9950/v1">
</div>
<div class="form-group">
<label for="sentenceStrategy">分句策略</label>
<select id="sentenceStrategy">
<option value="0" selected>短停顿800ms</option>
<option value="1">长停顿1400ms</option>
</select>
</div>
<!-- 展示合并与内部音频切段分开配置,便于对照原始片段。 -->
<label><input type="checkbox" id="displayMerge" checked> 合并相邻且已确认的同一说话人</label>
<div class="form-row">
<div class="form-group">
<label>话者分离</label>
<label class="toggle">
<input type="checkbox" id="speakerDiarization" checked>
<span class="toggle-slider"></span>
<span class="toggle-label" id="diarizationLabel">开启</span>
</label>
</div>
</div>
</div>
<div class="section-divider"></div>
<div class="audio-input-stack">
<div class="input-mode-tabs">
<button class="mode-tab active" id="tabMic" data-mode="mic">🎙️ 麦克风</button>
<button class="mode-tab" id="tabFile" data-mode="file">📁 文件</button>
</div>
<div class="input-mode-panel" id="panelMic">
<div class="mic-status" id="micStatus">点击下方按钮开始录音</div>
<div class="mic-timer" id="micTimer" style="display:none">🔴 录音中 <span id="micElapsed">00:00</span></div>
</div>
<div class="input-mode-panel" id="panelFile" style="display:none">
<div class="file-select">
<label class="btn btn-outline" for="audioFile">选择音频文件</label>
<input type="file" id="audioFile" accept=".pcm,.wav" hidden>
</div>
<span class="file-info" id="fileInfo">支持 16kHz、单声道、PCM16 的 PCM/WAV</span>
<div class="audio-meta" id="audioMeta" style="display:none">
<div class="audio-meta-row"><span class="meta-k">格式</span><span class="meta-v" id="metaFormat"></span></div>
<div class="audio-meta-row"><span class="meta-k">采样率</span><span class="meta-v" id="metaSampleRate"></span></div>
<div class="audio-meta-row"><span class="meta-k">时长</span><span class="meta-v" id="metaDuration"></span></div>
</div>
<div class="speed-control" id="speedControl" style="display:none">
<label class="speed-label">发送倍速 <span id="speedValue">1.0x</span></label>
<input type="range" id="speedSlider" min="0.5" max="3" step="0.1" value="1.0">
</div>
</div>
<div class="controls">
<button class="btn btn-primary" id="btnStart">▶ 开始识别</button>
<button class="btn btn-danger" id="btnStop" disabled>⏹ 停止</button>
</div>
</div>
</section>
</aside>
<!-- 右侧面板:识别结果 + 日志 -->
<main class="panel-right">
<section class="card result-card">
<h2>识别结果</h2>
<p id="speakerStatus" role="status">正在读取服务配置…</p>
<div class="result-meta" id="resultMeta" style="display:none">
<div class="meta-item">
<span class="meta-label">VoiceID:</span>
<span class="meta-value" id="voiceIdDisplay"></span>
<button class="btn-icon" id="btnCopyVoiceId" title="复制">📋</button>
<button class="btn btn-outline btn-sm btn-export" id="btnExportWav" title="导出音频" disabled>💾 导出WAV</button>
</div>
<div class="meta-item">
<span class="meta-label">状态:</span>
<span class="status-dot" id="statusDot"></span>
<span id="statusText">就绪</span>
</div>
</div>
<div class="result-area" id="resultArea">
<div class="placeholder" id="resultPlaceholder">选择音频来源并点击"开始识别"</div>
</div>
</section>
<section class="card log-card">
<div class="log-header">
<h2>ASR 原始日志</h2>
<button class="btn btn-outline btn-sm" id="btnClearLog">清空</button>
</div>
<div class="log-area" id="logArea"></div>
</section>
</main>
</div>
<!-- 版本号变更用于刷新浏览器缓存,确保加载“未知说话人”标签逻辑。 -->
<script src="app.js?v=213"></script>
</body>
</html>