@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
- yarn
- pnpm
- bun
npm i --save-exact @remotion/whisper-web@4.0.311
pnpm i @remotion/whisper-web@4.0.311
bun i @remotion/whisper-web@4.0.311
yarn --exact add @remotion/whisper-web@4.0.311
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.
Vite
To use @remotion/whisper-web
with Vite, you need to make two important changes:
- Exclude the package from Vite's dependency optimization to prevent known issues
- Configure the required security headers for
SharedArrayBuffer
support
vite.config.tstsx
import {defineConfig} from 'vite';export default defineConfig({optimizeDeps: {// turn off dependency optimization: https://github.com/vitejs/vite/issues/11672#issuecomment-1397855641exclude: ['@remotion/whisper-web'],},// required by SharedArrayBufferserver: {headers: {'Cross-Origin-Embedder-Policy': 'require-corp','Cross-Origin-Opener-Policy': 'same-origin',},},// ...});
Example usage
Transcribing with @remotion/whisper-webtsx
import {transcribe ,canUseWhisperWeb ,resampleTo16Khz ,downloadWhisperModel } from '@remotion/whisper-web';constfile = newFile ([], 'audio.wav');constmodelToUse = 'tiny.en';const {supported ,detailedReason } = awaitcanUseWhisperWeb (modelToUse );if (!supported ) {throw newError (`Whisper Web is not supported in this environment: ${detailedReason }`);}console .log ('Downloading model...');awaitdownloadWhisperModel ({model :modelToUse ,onProgress : ({progress }) =>console .log (`Downloading model (${Math .round (progress * 100)}%)...`),});console .log ('Resampling audio...');constchannelWaveform = awaitresampleTo16Khz ({file ,onProgress : (p ) =>console .log (`Resampling audio (${Math .round (p * 100)}%)...`),});console .log ('Transcribing...');const {transcription } = awaittranscribe ({channelWaveform ,model :modelToUse ,onProgress : (p ) =>console .log (`Transcribing (${Math .round (p * 100)}%)...`),});console .log (transcription .map ((t ) =>t .text ).join (' '));
Functions
canUseWhisperWeb()
Check if the browser supports this Whisper package
downloadWhisperModel()
Download a Whisper model in browser
resampleTo16khz()
Prepare an audio file for transcription by resampling it to 16kHz
transcribe()
Transcribe an audio file in the WASM environemnt
getLoadedModels()
Get a list of Whisper models that are already downloaded
toCaptions()
Convert the output from
transcribe()
to an array of Caption
objects.License
MIT