By Team Edtoolkits · Audio
Transcribe English speech from an audio or video file on your device with the open Whisper model, with timestamps and TXT, SRT or VTT export. The model downloads only when you ask.
Use the guide alongside the working tool.
Open Speech to Text Transcriber →Get started
Choose an audio or video file, or try the sample.
Adjust and explore
Press Start. The first time, the speech model downloads (about 41 MB) and is kept by your browser.
Save and share
Fix any words, click a time to hear it, then download TXT, SRT or VTT.
Privacy and limits
English only (the small English Whisper model). Accuracy drops with noise, strong accents and several people talking at once, and there is no speaker labelling. Time stamps are about one second accurate. It runs on your device, so long files are slow on phones; keep files under 60 minutes. Nothing is uploaded.
Ready to apply this workflow?
Try Speech to Text Transcriber →Explore a related workflow
- Record a video voiceover with timed takes and background music
Add continuous or interval-based narration to a video, edit takes, mix music and export a private browser-local MP4.
- Audio Loop Seams: Check the Shorter Cycle After Crossfading
Compare zero-crossing matching and tail-to-head blending while preserving awareness of loop duration.
- Choose between read-aloud and downloadable speech
Compare read-aloud voices with local audio export, check pronunciation and choose an appropriate privacy path for your text.
- Voice Recorder in your browser
Record your voice with a live level meter and noise reduction, pause and resume, trim the ends and save as MP3 or WAV. Nothing is uploaded.
Found unclear instructions or an unexpected result? Contact Team Edtoolkits.