Whisper transcription flow
1
Upload
Choose an audio or video file from your device.
2
Load Whisper
The browser loads the matching Whisper tiny model only when needed.
3
Export
Edit the transcript, copy it, or download TXT, SRT, and VTT files.
Whisper Online Transcription
Whisper audio to text in the browser
Use this Whisper-focused transcription page when you want a browser-based Whisper alternative for meeting notes, interviews, podcasts, lectures, and subtitle drafts.
Whisper tiny modelsCDN model loadingWorker-based transcriptionTXT / SRT / VTT export
Upload audio or video
Supports audio and common video files, max 100MB
Initializing upload...
A sponsor page opens in a new tab while processing continues here.
Upload audio or video to start Whisper transcription.
Transcript result
-- · 0 segments · 0 words
PDF opens in your browser print dialog and uses the edited text above.
Better Whisper results
- For noisy recordings, denoise a short section first and compare the Whisper transcript.
- Use fast English recognition for English-only audio, and multilingual mode for other languages.
- For long recordings, test a short section before processing the full file.
Whisper transcription FAQ
Is this a Whisper alternative?
Yes. It uses Whisper-style browser transcription with Xenova Whisper tiny models, so you can test audio to text without sending the source file to Resodock servers.
Why does Whisper download files first?
The browser needs model assets before it can transcribe. They are loaded on demand from the CDN or proxy and are cached by the browser when possible.
Which Whisper mode should I choose?
Use fast English recognition for English recordings. Choose multilingual recognition for Chinese, Japanese, Korean, Spanish, French, German, or mixed-language audio.