Yap Audio to Text turns an MP3, M4A, WAV or video file into editable text, plus SRT or VTT subtitle files with real start and end times. OpenAI's Whisper speech model runs directly in your browser, so the file is never uploaded to a server and nothing is stored on our side. Files can be up to 60 minutes and 300 MB. Choose English mode or an any-language mode with 99 languages, detected automatically and never translated. Long recordings are cut at natural pauses so each subtitle has sensible timing. Edit the transcript in the page, copy it, or save a .txt file. Edits made in the text box do not carry into the SRT or VTT files. The first run downloads a speech model of about 80 MB, or about 165 MB with GPU acceleration. There is no sign-up and no minute limit.
We already make Yap, a voice dictation app for Mac and Windows that keeps speech on the device. The same privacy idea works for recordings, so we built a free browser tool for transcripts and subtitles.