|
<html lang="en"> |
|
|
|
<head> |
|
<meta charset="utf-8"> |
|
<meta name="viewport" content="width=device-width" /> |
|
<title>Next-gen Kaldi WebAssembly with sherpa-onnx for VAD + ASR</title> |
|
<style> |
|
h1,div { |
|
text-align: center; |
|
} |
|
textarea { |
|
width:100%; |
|
} |
|
.loading { |
|
display: none !important; |
|
} |
|
</style> |
|
</head> |
|
|
|
<body style="font-family: 'Source Sans Pro', sans-serif; background-color: #f9fafb; color: #333; display: flex; flex-direction: column; align-items: center; height: 100vh; margin: 0;"> |
|
<h1> |
|
Next-gen Kaldi + WebAssembly<br/> |
|
VAD+ASR Demo with <a href="https://github.com/k2-fsa/sherpa-onnx">sherpa-onnx</a><br/> |
|
(with Zipformer supporting Chinese 中文) |
|
</h1> |
|
|
|
<div style="width: 100%; max-width: 900px; background: #fff; padding: 1.5rem; border-radius: 8px; box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1); flex: 1;"> |
|
<div id="status">Loading...</div> |
|
|
|
<div id="singleAudioContent" class="tab-content loading"> |
|
<div style="display: flex; gap: 1.5rem;"> |
|
<div style="flex: 1; display: flex; flex-direction: row; align-items: center; gap: 1rem;"> |
|
<button id="startBtn" disabled>Start</button> |
|
<button id="stopBtn" disabled>Stop</button> |
|
<button id="clearBtn">Clear</button> |
|
</div> |
|
</div> |
|
|
|
<div style="flex: 1; display: flex; flex-direction: column; gap: 1rem;"> |
|
<div style="font-size: 1rem; font-weight: bold; padding: 0.5rem 1rem; background-color: #f8f9fa; border-radius: 8px; color: #6c757d;">Transcript</div> |
|
<textarea id="results" rows="10" placeholder="Output will appear here..." readonly style="flex: 1; padding: 0.75rem; font-size: 1rem; border: 1px solid #ced4da; border-radius: 8px; resize: none; background-color: #f8f9fa;"></textarea> |
|
</div> |
|
|
|
<section flex="1" overflow="auto" id="sound-clips"> |
|
</section> |
|
</div> |
|
|
|
|
|
<div style="width: 100%; max-width: 900px; margin-top: 1.5rem; background: #fff; padding: 1.5rem; border-radius: 8px; box-shadow: 0 2px 8px rgba(0, 0, 0, 0.1); text-align: left; font-size: 0.9rem; color: #6c757d;"> |
|
<h3>Description</h3> |
|
<ul> |
|
<li>Everything is <strong>open-sourced.</strong> <a href="https://github.com/k2-fsa/sherpa-onnx">code</a></li> |
|
<li>If you have any issues, please either <a href="https://github.com/k2-fsa/sherpa-onnx/issues">file a ticket</a> or contact us via</li> |
|
<ul> |
|
<li><a href="https://k2-fsa.github.io/sherpa/social-groups.html#wechat">WeChat group</a></li> |
|
<li><a href="https://k2-fsa.github.io/sherpa/social-groups.html#qq">QQ group</a></li> |
|
<li><a href="https://k2-fsa.github.io/sherpa/social-groups.html#bilibili-b">Bilibili</a></li> |
|
</ul> |
|
</ul> |
|
<h3>About This Demo</h3> |
|
<ul> |
|
<li><strong>Private and Secure:</strong> All processing is done locally on your device (CPU) within your browser with a single thread. No server is involved, ensuring privacy and security. You can disconnect from the Internet once this page is loaded.</li> |
|
<li><strong>Efficient Resource Usage:</strong> No GPU is required, leaving system resources available for webLLM analysis.</li> |
|
</ul> |
|
<h3>Latest Update</h3> |
|
<ul> |
|
<li>Update UI.</li> |
|
<li>First working version.</li> |
|
</ul> |
|
|
|
<h3>Acknowledgement</h3> |
|
<ul> |
|
<li>We refer to <a href="https://huggingface.co/spaces/Banafo/Kroko-Streaming-ASR-Wasm">https://huggingface.co/spaces/Banafo/Kroko-Streaming-ASR-Wasm</a> for the UI part.</li> |
|
</ul> |
|
</div> |
|
|
|
<script src="sherpa-onnx-asr.js"></script> |
|
<script src="sherpa-onnx-vad.js"></script> |
|
<script src="app-vad-asr.js"></script> |
|
<script src="sherpa-onnx-wasm-main-vad-asr.js"></script> |
|
</body> |
|
|
