You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
154 lines
5.4 KiB
154 lines
5.4 KiB
2 years ago
|
<!doctype html>
|
||
|
<html lang="en-us">
|
||
|
<head>
|
||
|
<title>whisper.cpp : WASM example</title>
|
||
|
</head>
|
||
|
<body>
|
||
|
<div id="main-container">
|
||
|
Minimal <b>whisper.cpp</b> example using Javascript bindings
|
||
|
|
||
|
<br><br>
|
||
|
|
||
|
Model:
|
||
|
<input type="file" id="file" name="file" onchange="loadFile(event, 'ggml.bin')" />
|
||
|
<br><br>
|
||
|
|
||
|
WAV:
|
||
|
<input type="file" id="file" name="file" onchange="loadAudio(event)" />
|
||
|
<br><br>
|
||
|
|
||
|
<button onclick="onTranscribe();">Transcribe</button>
|
||
|
|
||
|
<br><br>
|
||
|
|
||
|
<div class="cell-version">
|
||
|
<span>
|
||
|
|
|
||
|
Build time: <span class="nav-link">@GIT_DATE@</span> |
|
||
|
Commit hash: <a class="nav-link" href="https://github.com/ggerganov/whisper.cpp/commit/@GIT_SHA1@">@GIT_SHA1@</a> |
|
||
|
Commit subject: <span class="nav-link">@GIT_COMMIT_SUBJECT@</span> |
|
||
|
<a class="nav-link" href="https://github.com/ggerganov/whisper.cpp/tree/master/examples/whisper.wasm">Source Code</a> |
|
||
|
</span>
|
||
|
</div>
|
||
|
</div>
|
||
|
|
||
|
<script type="text/javascript" src="whisper.js"></script>
|
||
|
<script type='text/javascript'>
|
||
|
window.AudioContext = window.AudioContext || window.webkitAudioContext;
|
||
|
window.OfflineAudioContext = window.OfflineAudioContext || window.webkitOfflineAudioContext;
|
||
|
|
||
|
// web audio context
|
||
|
var context = null;
|
||
|
|
||
|
// audio data
|
||
|
var audio = null;
|
||
|
|
||
|
// the whisper module instance
|
||
|
var whisper = null;
|
||
|
var instance = null;
|
||
|
|
||
|
// instantiate the whisper instance
|
||
|
// whisper_factory comes from the whisper.js module
|
||
|
whisper_factory().then(function(obj) {
|
||
|
whisper = obj;
|
||
|
});
|
||
|
|
||
|
// helper function
|
||
|
function convertTypedArray(src, type) {
|
||
|
var buffer = new ArrayBuffer(src.byteLength);
|
||
|
var baseView = new src.constructor(buffer).set(src);
|
||
|
return new type(buffer);
|
||
|
}
|
||
|
|
||
|
// initialize whisper
|
||
|
function init() {
|
||
|
if (!instance) {
|
||
|
instance = whisper.init('ggml.bin');
|
||
|
if (instance) {
|
||
|
console.log('whisper instance initialized');
|
||
|
}
|
||
|
}
|
||
|
|
||
|
if (!instance) {
|
||
|
console.log('whisper instance initialization failed');
|
||
|
return;
|
||
|
}
|
||
|
|
||
|
if (instance) {
|
||
|
var ret = whisper.full_default(instance, audio);
|
||
|
if (ret) {
|
||
|
console.log('whisper full_default returned: ' + ret);
|
||
|
}
|
||
|
}
|
||
|
}
|
||
|
|
||
|
function loadFile(event, fname) {
|
||
|
var file = event.target.files[0] || null;
|
||
|
if (file == null) {
|
||
|
return;
|
||
|
}
|
||
|
|
||
|
console.log(
|
||
|
"<p>File information: <strong>" + file.name +
|
||
|
"</strong> type: <strong>" + file.type +
|
||
|
"</strong> size: <strong>" + file.size +
|
||
|
"</strong> bytes</p>"
|
||
|
);
|
||
|
|
||
|
var reader = new FileReader();
|
||
|
reader.onload = function(event) {
|
||
|
var buf = new Uint8Array(reader.result);
|
||
|
|
||
|
// write to WASM file using whisper.FS_createDataFile
|
||
|
whisper.FS_createDataFile("/", fname, buf, true, true);
|
||
|
}
|
||
|
reader.readAsArrayBuffer(file);
|
||
|
}
|
||
|
|
||
|
function loadAudio(event) {
|
||
|
if (!context) {
|
||
|
context = new AudioContext({sampleRate: 16000});
|
||
|
}
|
||
|
|
||
|
var file = event.target.files[0] || null;
|
||
|
if (file == null) {
|
||
|
return;
|
||
|
}
|
||
|
|
||
|
console.log(
|
||
|
"<p>Audio information: <strong>" + file.name +
|
||
|
"</strong> type: <strong>" + file.type +
|
||
|
"</strong> size: <strong>" + file.size +
|
||
|
"</strong> bytes</p>"
|
||
|
);
|
||
|
|
||
|
var reader = new FileReader();
|
||
|
reader.onload = function(event) {
|
||
|
var buf = new Uint8Array(reader.result);
|
||
|
|
||
|
context.decodeAudioData(buf.buffer, function(audioBuffer) {
|
||
|
var offlineContext = new OfflineAudioContext(audioBuffer.numberOfChannels, audioBuffer.length, audioBuffer.sampleRate);
|
||
|
var source = offlineContext.createBufferSource();
|
||
|
source.buffer = audioBuffer;
|
||
|
source.connect(offlineContext.destination);
|
||
|
source.start(0);
|
||
|
|
||
|
offlineContext.startRendering().then(function(renderedBuffer) {
|
||
|
audio = renderedBuffer.getChannelData(0);
|
||
|
//var audio16 = convertTypedArray(data, Int16Array);
|
||
|
});
|
||
|
});
|
||
|
}
|
||
|
reader.readAsArrayBuffer(file);
|
||
|
}
|
||
|
//
|
||
|
// Transcribe
|
||
|
//
|
||
|
|
||
|
function onTranscribe() {
|
||
|
init();
|
||
|
}
|
||
|
</script>
|
||
|
</body>
|
||
|
</html>
|