Transcribe a recording
Upload an audio or video file you own. The free tier includes 60 minutes a month after a free sign-up.
Open the transcriberFree tools
Two free ways to work with a recording: clean up an SRT or VTT file right here, or transcribe a file you own. The converter below runs entirely in this tab — your subtitle text never leaves the browser.
Paste SRT or VTT, or open a file from your device. Choose an output format, optionally shift the timing, then copy or download.
Negative moves earlier; 0 keeps the original timing.
Runs locally in this browser. No upload, no server, no account.
The converter is the fast path. When you need the words in the first place, these are the next steps.
Upload an audio or video file you own. The free tier includes 60 minutes a month after a free sign-up.
Open the transcriberBrowse by file format and use case — MP3, WAV, M4A, MP4, audio to SRT, video to text, and more.
Browse all toolsSee the exact workflow before you upload: audio to SRT, MP4 to text, video to text.
Read a guideDrop in the subtitle text you already have, or open an .srt / .vtt file from your device.
SRT and VTT are interchangeable; TXT strips timestamps for show notes or an article draft.
Fix the timing first if needed. Nothing is saved on a server, so keep your own copy.
No. Parsing, conversion, and download all run in your browser. The text is never sent to Dictawave or a third party, and there is no account or tracking requirement for the converter.
SRT, WebVTT (VTT), and plain text. SRT and VTT keep the cue timing; TXT removes timestamps and styling tags for a readable transcript. You can also shift every cue earlier or later by a number of seconds.
No. The converter works without signing in. Transcription requires a free account; the free tier includes 60 minutes a month.
No. Dictawave does not download or extract content from third-party platforms. The converter only processes text you paste or a file you open from your own device.
If you need ASS, TTML, or a different timecode convention, tell us. We add formats based on real transcription work, not feature lists.