Whisper Web
A free multilingual speech-to-text tool that runs locally in the browser
About Whisper Web
Whisper Web is a browser-based speech-to-text tool built on OpenAI Whisper, with automatic recognition for more than 100 languages. Free mode processes audio locally on the device with WebGPU and WebAssembly, so audio never leaves the browser; it handles files up to 200 MB and 20 minutes and exports TXT, SRT, VTT, or JSON. For longer files or batch uploads, choose the cloud-based Unlimited plan.
Best for
Best use cases
Local offline transcription
Multilingual audio recognition
Subtitle and text file export
Batch cloud transcription of long audio
Features
Local transcription
Free mode processes audio in the browser with WebGPU and WebAssembly.
Automatic multilingual detection
Supports more than 100 languages, with AI automatically detecting the spoken language.
Multi-format export
Transcriptions can be exported as TXT, SRT, VTT, or JSON.
Long cloud audio
Unlimited supports files up to 10 hours and 5 GB each, with up to 50 files uploaded at once.
Pros
- Free mode processes audio locally in the browser, so it never leaves the device
- Supports more than 100 languages and detects the language automatically
- Exports to TXT, SRT, VTT, and JSON
- No account registration required
Cons
- Free mode limits each file to 200 MB and 20 minutes
- Unlimited cloud features require an internet connection
Pricing Plans
Free
- Maximum 200 MB and 20 minutes per file
- Automatic detection for more than 100 languages
- Local processing with TXT, SRT, VTT, and JSON export
Unlimited
- Unlimited cloud transcription
- No device limit
- Up to 10 hours and 5 GB per file
- Upload 50 files at once
Unlimited Annual
- Unlimited cloud transcription
- No device limit
- Up to 10 hours and 5 GB per file
- Upload 50 files at once
* Pricing information is for reference only. Please refer to the latest announcement on Whisper Web's official website.