1. Upload your file
Open the Audio to SRT converter and drop an MP4, MP3 or WAV file on the upload area (or click it to browse). Nothing is uploaded: the file is decoded and transcribed in your browser's memory, on your own device. The first run downloads the speech model once; after that it is cached.
2. Choose the language
Leave Auto detect on, or pick the spoken language from the list of 99 languages. Choosing it yourself helps with short clips and mixed-language audio. Fast quality works on any computer; Accurate uses a larger model and is best with a recent browser that supports WebGPU.
3. Pick a caption style
- Vertical — short, punchy cues for TikTok, Reels and Shorts. Start with 2–4 words per cue.
- Horizontal — classic subtitles for YouTube, courses and films. 6–10 words per cue reads comfortably.
- Per word — one word per cue, for karaoke and kinetic typography.
The Max words per cue slider sets the longest a cue can get; cues also break at sentence ends and pauses.
4. Preview and download
Press Generate SRT. When it is done, play the preview to check the timing, fix any word in the text box, then Download .srt (or .vtt). Changing the style or the slider re-cuts the captions instantly — no need to transcribe again.
Tip: import the .srt into Premiere Pro, DaVinci Resolve, Final Cut Pro or CapCut as a caption track, then style it there.