Private Whisper Transcriber

Single-file browser app for local audio transcription with Transformers.js and Whisper. Results are stored in your browser’s IndexedDB and can be exported as plain text or Word-compatible .docx.

No audio upload: the selected audio file is decoded and transcribed in this browser tab. On first use, the browser downloads JavaScript libraries, ONNX runtime assets, and the selected Whisper model from public CDNs/model hosting; your audio and transcript are not sent to a server by this app.

1. Audio and model

Audio file Whisper model

Device Data type

Task Language hint

Chunk length, seconds Stride/overlap, seconds

Include chunk timestamps in the saved/exported transcript

Choose an audio file, then load the model or start transcription.

For best stability, start with tiny.en or base.en. Larger models need more memory and may be slow on CPU-only browsers.

1. Audio and model

2. Transcript

3. IndexedDB history