Audio to Text Transcription
Audio to text transcription turns spoken words in a recording into written text. Upload a voice memo, interview, meeting, lecture or podcast, and AI speech recognition writes out everything that is said, with timestamps.
Upload an audio or video file up to 2 GB and 8 hours long. Get a timestamped transcript in 99 languages, then summarize it, chat with it or export it.
99 languages
Detects the language automatically, and can translate the transcript as it goes.
Long files welcome
Files up to 2 GB and 8 hours: full meetings, lectures and podcast episodes.
Speaker labels
Optionally label who is speaking, for interviews and meetings.
HOW TO CONVERT AUDIO TO TEXT
Choose speech-to-text
In the Extract box, pick “Speech to text only”.
Upload your file
Drop in an MP3, WAV, M4A, MP4, WebM or other common audio or video file.
Set options if you need them
Choose the language, turn on speaker labels, or add names and jargon to improve accuracy.
Read, export or analyze
Download TXT, SRT, VTT or JSON, or summarize and chat with the transcript.
WHAT PEOPLE TRANSCRIBE
- ✓Interviews and research calls, ready to quote and analyze
- ✓Meeting recordings, turned into notes and action items
- ✓Lectures and classes, turned into study notes
- ✓Podcast episodes, turned into show notes and blog posts
- ✓Voice memos and dictation, turned into written drafts
- ✓Videos for subtitles in SRT or VTT
FREQUENTLY ASKED QUESTIONS
How do I transcribe an audio file to text?
Sign in, pick “Speech to text only” in the Extract box, upload your file and click Extract. The transcript opens with timestamps, ready to read or export.
What audio and video formats can I upload?
Common audio and video files such as MP3, WAV, M4A, MP4 and WebM, up to 2 GB and 8 hours long.
How much does audio transcription cost?
1 credit per 2 minutes of audio. Every account gets 10 free credits a month, and pay-as-you-go credit packs never expire. Some advanced options, such as speaker labels, add a small per-hour cost.
How accurate is it?
Clear recordings come out very accurate. For names, products or jargon, add them as key terms (English audio) to improve recognition.
Can it tell different speakers apart?
Yes. Turn on speaker labels and each line is marked with who said it. You can also map speakers to names or roles.
Can I translate the transcript?
Yes. Choose a language under “Translate to” before you start, or translate a finished transcript into 100+ languages.