TRANSCRIBER / GOOD TO KNOW
Audio to text with timestamps
Upload an audio file, wait for recognition, then edit the passages that need attention. Recordings with several participants include editable speaker names.
Choose your source file
Common formats include MP3, M4A, WAV, FLAC, OGG, and OPUS. We inspect the actual file contents, so changing an extension cannot make an unsupported recording compatible.
Use the original recording when possible. Repeated compression can reduce intelligibility, and simply making a recording louder does not restore lost speech.
Select a language
Qwen3-ASR supports 30 languages. Automatic detection works for ordinary recordings; choose a language manually when a short clip or specialist vocabulary makes detection difficult.
Segment timestamps are available across supported languages. Word alignment depends on the language; check timestamp accuracy and speaker changes against the source.
Edit and download
Correct the transcript and save it as TXT or DOCX. Editing and export run in your browser, so changes are not sent back to the server.
The temporary audio copy is deleted after successful processing. The server deletes the transcript after the browser confirms delivery and shows a separate deletion confirmation. Download your copy before refreshing the tab.