xenova

whisper-web: Run Whisper Speech Recognition in Your Browser

Whisper Web runs speech recognition directly in your browser using Transformers.js, the JavaScript library from Hugging Face. It is a browser-based option for trying Whisper transcription without setting up a separate recognition service.

Stars
3.3k
Forks
433
Stars, last 30 days
+0
Last commit
2.1 years ago
Contributors
3

Our take

Inactiveno commits in over a year

Health score

22/100

How it is scored
Activity
19
Community
26
Issues
21
Pull requests
20

+0 stars in the last 30 days

  • Fewer contributors than 82% of the projects we track

Whisper Web is a focused browser demo and prototype, but its inactivity and concentrated maintenance make it a cautious choice for production use.

Good fit if

  • You want to explore Whisper transcription in a browser using Transformers.js.
  • You are comfortable running a TypeScript app with Node.js and npm.
  • You can evaluate and maintain the code yourself if you build on it.

Look elsewhere if

  • You need an actively maintained project: there have been no commits in over a year and no releases in the past year.
  • You need an established maintenance base: most of the code comes from one person, and no issues or pull requests were closed or merged recently.
  • You require a clearly stated license before adopting or redistributing the code; the supplied repository data does not identify one.
All health signals
Last commit2024-10-01 (2.1 years ago)
Commits, last 90 days0
Releases, last 12 months0
Contributors3 (top contributor: 84% of commits)
Issues closed, last 90 days0
Pull requests merged, last 90 days0
Project age3 years

Checked on 2026-11-02 with the GitHub API.

Project history

Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.

Selected snapshot

Stars

3,345

Forks

431

Change since Nov 17, 2026: 0 stars·0 forks

Overview

Whisper Web brings Whisper speech recognition to the browser, using Transformers.js from Hugging Face. It is aimed at people who want to transcribe speech in a browser rather than build or operate a separate recognition service.

The project is a TypeScript web app with a hosted Hugging Face demo and a local development setup. Its browser-based approach makes it useful for experimentation and lightweight transcription workflows, though its current maintenance status matters if you plan to depend on it in a larger product.

Key Features

  • Runs speech recognition directly in the browser.
  • Uses Transformers.js for its machine-learning implementation.
  • Provides a hosted demo on Hugging Face Spaces.
  • Can be run locally with Node.js and npm.
  • Uses browser Web Workers; Firefox users may need to enable module workers in browser settings.

Use Cases

  • Developers evaluating browser-based Whisper transcription before building a speech feature.
  • Users who want to try speech recognition through the hosted demo without installing a local development environment.
  • Frontend developers prototyping a transcription interface that runs in the browser.
  • Teams assessing whether Transformers.js fits a browser-based machine-learning workflow.

What you need

Detected in the repository

  • Node.js (from package.json)

Getting Started

git clone https://github.com/xenova/whisper-web.git
cd whisper-web
npm install
npm run dev

See the repository README for additional setup details, including the Firefox Web Workers setting.

License in Plain Words

No standard license was detected. Without a license, the default copyright rules apply: you may read the code, but not reuse it. Check the repository or ask the authors.

Alternatives

  • insanely-fast-whisper: It transcribes audio from a terminal using GPU-oriented tooling, rather than running Whisper in a browser.
  • whisper.cpp: It provides a lightweight C/C++ Whisper implementation for offline apps and services, rather than a browser-based Transformers.js app.
  • meeting-transcriber: It records and transcribes online meetings in a macOS menu bar app, rather than transcribing audio in a browser.
ProjectLanguageLicenseStarsStatus
whisper-webTypeScript–3.3kInactive
insanely-fast-whisperJupyter NotebookApache-2.013.1kQuiet
whisper.cppC++MIT54.1kActive
meeting-transcriberSwiftMIT222Active

Considerations

The repository is marked inactive, with no recent commits or releases, and contributor activity is concentrated in one person. The supplied requirements identify Node.js, while the README notes a Firefox setting that may need changing for Web Workers. No test setup or CI is indicated in the supplied project facts, and no license is listed in the supplied data, so verify the repository's licensing terms before using or distributing it.

Found this useful?

Share it with someone who would like whisper-web.

Comparisons