Turn interviews, meetings, lectures, voice memos and videos into text with Whisper AI — running on your own device. Export plain text or SRT / VTT subtitles.
Drag & drop an audio or video file here
or browse your files
MP3, WAV, M4A, MP4, MOV, WebM… · Max 100 MB per file
The transcript will appear here. Download it as text (TXT) or subtitles (SRT, VTT).
A transcript makes a recording searchable, quotable and accessible: interview notes, meeting minutes, lecture notes, podcast show notes, subtitles for deaf and hard-of-hearing viewers, or simply reading instead of listening. Text is also what search engines index.
Whisper is the open speech recognition model released by OpenAI and trained on 680,000 hours of audio in about a hundred languages. Here it runs locally through Transformers.js: the sound is decoded by FFmpeg, cut into 30-second windows at quiet moments, and transcribed on your processor or graphics card.
The TXT export gives clean paragraphs, a new one after each pause. SRT and VTT keep the timing of every sentence, cut into readable two-line subtitles, ready for YouTube, VLC, Premiere or a web player.
Drop an audio or video file: MP3, WAV, M4A, FLAC, MP4, MOV, WebM…
Choose the language spoken, Fast or Precise, and optionally the part to transcribe.
Copy the text, or download it as TXT, SRT or VTT subtitles.
Record close to the speaker, in a quiet room: the cleaner the sound, the better the text. A phone placed in front of the speaker beats a laptop microphone across the room.
Long meeting? Transcribe the part you need with the handles: it is faster and you avoid hours of small talk.
Pro tipThe speech model runs in your browser: your recording is never uploaded, stored or listened to by anyone.
Safe for interviews, medical or legal recordings, meetings and anything confidential.
PrivacyEverything about audio transcription