Transcribe audio to text online

How to transcribe a recording

1. Upload an audio or video file. Only the sound is taken from a video, and the original file is not kept.
2. Pick the language or leave it on Auto.
3. Add names and terms that appear in the recording: people, brands, jargon. The list is a hint for recognition.
4. Press Transcribe. Long recordings run in the background: you can close the tab, we email you when the text is ready, and it stays under My transcripts.

What you get

• Text in paragraphs with timecodes like [12:31].
• A player next to the text. Click a word or a timecode to jump there; the current word is highlighted while it plays.
• Search and editing right on the page, and Copy all in one click.
• Downloads: TXT with or without timecodes, DOCX for Word or Google Docs, SRT and VTT subtitles.
• In brief, topics and action items, written automatically after recognition: three lines with the essence, topics with timecodes (click one to jump to that place) and who does what by when, if it was said. After sign-in you can also get a report in another format: meeting, interview, lecture, podcast or your own request.

SRT and VTT subtitles for a video

Subtitles are built from the same transcript with a time for every word, so you do not need a separate tool. Each cue has up to two lines of up to 42 characters and stays on screen for 1 to 7 seconds, with breaks at pauses rather than in the middle of a phrase. Upload the SRT file to YouTube, or import SRT or VTT into a video editor.

If you fix a word in the text, the paragraph keeps its timing and the words inside it are spread again. Step-by-step guide: how to make SRT subtitles from a video for free.

Voice messages and voice notes to text

Voice messages saved from messengers usually come as OGG or OPUS files, and phone voice memos as M4A or AMR. All of them are accepted as they are, no conversion needed. A short voice note fits easily into the free minutes, even without an account.

Meetings, interviews and lectures

Upload the recording from your call or recorder. Right after recognition the app writes a three-line brief, topics with timecodes and the action items that were named. After sign-in, Another report from the text turns the transcript into one of five formats: meeting, interview, lecture, podcast, or your own request, such as "list every decision and who owns it". Guide: how to transcribe a meeting or an interview.

Ask the participants for consent before you record or transcribe a call. Speaker labels are not in the first version, so in a dialogue the text comes as paragraphs split by pauses.

Languages

Files and limits

Privacy

The audio is deleted right after transcription. The text is kept for 30 days, and Delete removes it at once. Recordings are not used to train models. The brief, topics and action items are written by an AI service from the finished text: it receives only the transcript text, without the audio or account data.

Tips for a cleaner transcript

• Music under the voice makes recognition harder. If you have a version without music, upload that. For a song, split the voice from the music first with the vocal remover.
• Noisy or echoey interviews read better after cleanup: the interview set in audio restoration reduces hiss, hum and room echo.
• Only need a part? Cut it out first with the audio cutter and save your free minutes.
• Overlapping voices may leave gaps in those places.
• Rare names and terms are best added to the list before you start.

Related tools

• Podcast mastering: even loudness for a podcast episode before you publish it with its transcript.
• Music visualizer: a beat-synced video for a song, ready for TikTok, Reels and Shorts.
• Vocal remover: the vocal and the instrumental of a song as two files.
• Audio restoration: cassette, vinyl, phone and interview cleanup.