Turn audio into text
Upload a recording and get a transcript you can search, review, translate, and download.
- Maximum 30 minute duration, basic accuracy
- Maximum 3 uploads a day when signed in
Start to finish
What you get
Export video with subtitlesPreview and adjust your subtitles, then export one free watermarked clip of up to five minutes.
Translate your transcriptionUse included Basic translations or Pro models with credit.
Downloads & ActionsDownload SRT, VTT or TXT subtitle files, or open the transcription in the player.
Transcript editorProCorrect the transcript line by line while every subtitle file stays in sync.
Ask this videoGet answers without scrubbing through the file. Free uploads include a limited number of questions.
Highest accuracyProThe best transcription models available, not the fastest ones — so there is less to fix afterwards.
Searchable transcriptSearch every word, filter by speaker, and jump straight to the moment anything was said.
Media PlayerWatch your file with the subtitles over it, switch between translated tracks, and set their size, spacing and background.
Video ImporterPaste a URL and we fetch the media for you from 1,100+ sites.
Know who is speakingProSpeaker labels on every line, with names added automatically.
Whole batches at onceProDrop in many files and let them transcribe in parallel.
API accessProConnect your own software with an API key.
$0.99/ hour
Pay only for the hours you actually process, and every hour runs on the best transcription models available.
- No subscription
- Start with $0.99
- Credit never expires
Recordings hold useful conversations and details, yet finding one sentence inside an hour of audio can take a long time. Upload an interview, lecture, call, meeting, sermon, research recording, podcast, or voice memo here. The result is a timed transcript you can read on screen, check against the recording, translate when needed, and download in a format that fits the next step.
How to do it
- Upload an audio file such as MP3, WAV, M4A, AAC, OGG, or FLAC.
- Leave the language on auto-detect, or choose it when you already know it.
- Read the transcript, search it, and correct any names or specialist terms.
- Download TXT for reading, or SRT, VTT, or ASS when you need timing.
Questions
Can I start without creating an account?
Yes. Choose the free option in the uploader. It shows the current file limit and whether a free use is available before you start.
Can the transcript show who is speaking?
Yes. Choose Highest quality and turn on Identify Speakers for recordings with several voices. You can rename each speaker in the transcript.
What helps a difficult recording?
Clear speech and a microphone close to the speaker help most. For echo, background noise, quiet voices, or several people talking, choose the highest available quality and review names carefully.
Which download should I choose?
TXT is easiest to read and quote. SRT, VTT, and ASS keep timing so the words can stay connected to the recording.
Free use, prices, and limits
These numbers are the ones the uploader on this page actually enforces.
- Starting out
- No account or payment details are needed for the first free upload. A free account lets you upload again immediately, then every 30 minutes, up to 3 uploads per 24 hours.
- Free transcriptions
- 1 per 24 hours without an account · up to 3 per 24 hours with a free account, one every 30 minutes
- Free file length
- Up to 30 minutes per file without an account · up to 30 minutes with a free account
- Free file size
- Up to 1 GB per upload
- Pro price
- $0.99 per hour of audio or video, paid as you go with no subscription. The first purchase starts at $0.99; later top-ups start at $5.00.
- Pro file limits
- No duration limit. Files up to 10 GB.
- Downloads
- Transcript and subtitle files (TXT, SRT, VTT with every transcription, ASS with the paid option), or a downloadable video with the subtitles included
- Languages
- Transcription in 100 languages with automatic detection · translation into 120 languages
- Speaker identification
- Available with the paid Highest accuracy option. Detected speakers can be renamed in the transcript.