Skip to main content

@remotion/whisper-web

warning

Unstable API: This package is experimental for the moment. As we test it, we might make a few changes to the API and switch to a WebGPU-based backend in the future.

Similar to @remotion/install-whisper-cpp but for the browser. Allows you to transcribe audio locally in the browser, with the help of WASM.

Installation

npm i --save-exact @remotion/whisper-web@4.0.311
This assumes you are currently using v4.0.311 of Remotion.
Also update remotion and all `@remotion/*` packages to the same version.
Remove all ^ character in front of the version numbers of it as it can lead to a version conflict.

Required configuration

@remotion/whisper-web uses a WebAssembly (WASM) backend that requires SharedArrayBuffer support. To enable this functionality, you need to configure cross-origin isolation headers in your application. This is a security requirement for using SharedArrayBuffer in modern browsers.

See: Important Considerations

Vite

To use @remotion/whisper-web with Vite, you need to make two important changes:

  1. Exclude the package from Vite's dependency optimization to prevent known issues
  2. Configure the required security headers for SharedArrayBuffer support
vite.config.ts
tsx
import {defineConfig} from 'vite';
export default defineConfig({
optimizeDeps: {
// turn off dependency optimization: https://github.com/vitejs/vite/issues/11672#issuecomment-1397855641
exclude: ['@remotion/whisper-web'],
},
// required by SharedArrayBuffer
server: {
headers: {
'Cross-Origin-Embedder-Policy': 'require-corp',
'Cross-Origin-Opener-Policy': 'same-origin',
},
},
// ...
});

Example usage

Transcribing with @remotion/whisper-web
tsx
import {transcribe, canUseWhisperWeb, resampleTo16Khz, downloadWhisperModel} from '@remotion/whisper-web';
 
const file = new File([], 'audio.wav');
const modelToUse = 'tiny.en';
 
const {supported, detailedReason} = await canUseWhisperWeb(modelToUse);
if (!supported) {
throw new Error(`Whisper Web is not supported in this environment: ${detailedReason}`);
}
 
console.log('Downloading model...');
await downloadWhisperModel({
model: modelToUse,
onProgress: ({progress}) => console.log(`Downloading model (${Math.round(progress * 100)}%)...`),
});
 
console.log('Resampling audio...');
const channelWaveform = await resampleTo16Khz({
file,
onProgress: (p) => console.log(`Resampling audio (${Math.round(p * 100)}%)...`),
});
 
console.log('Transcribing...');
const {transcription} = await transcribe({
channelWaveform,
model: modelToUse,
onProgress: (p) => console.log(`Transcribing (${Math.round(p * 100)}%)...`),
});
 
console.log(transcription.map((t) => t.text).join(' '));

Functions

License

MIT