Speak into your mic and watch words appear — fully on-device.
Grant permission when your browser prompts.
Words appear in rolling windows as you talk.
Copy or download the final text.
It runs Whisper locally, so results are consistent across browsers and never leave your device.
Yes, once the Whisper model is cached.
Whichever microphone your browser selects — you can pick another from your OS settings.
Transcribe MP3, WAV, M4A, OGG or FLAC into text with Whisper.
Generate SRT and WebVTT captions for any video file.
Long-form transcript with timestamps for episodes and interviews.
Record a meeting and transcribe with headings and bullet notes.
Dictate a quick note and save it as a .md file.
Whisper translate mode — convert any language into English.
Identify the spoken language of any short audio clip.
Word-level timestamps aligned to the audio — perfect for karaoke.
Zro7