Transcribe audio to text.
Drop a file, read it, export it.
An audio to text converter for interviews, voice memos, lectures and video. Timestamps on every segment, editable before you copy or download TXT, SRT or VTT.
10 free minutes a day
MP3, WAV, M4A, MP4, MOV and more, up to 25 MB. Your browser extracts and compresses the audio first; the transcript comes back with timestamps.
Choose a fileAudio transcription without the project
No workspace to create, no team to invite, no "upload complete, check back later" email. The file goes in; the words come out on the same page.
Audio or video. Your browser pulls out the audio, compresses it and sends only that.
Segments with timestamps. Click a line to correct a name. Everything you change goes into the exports.
Plain text for notes and articles, SRT or VTT for subtitles, or copy with timestamps for show notes.
When you need an audio to text converter
Transcribing audio to text is the step between recording something and being able to use it. Interviews become quotes. Lectures become notes you can search. A voice memo recorded in the car becomes the first draft of a post. A talking-head video becomes captions, which is the difference between people watching with the sound off and scrolling past.
This audio transcription tool is deliberately plain: one drop zone, one transcript, three file formats. It handles the common formats directly and reads the audio track out of video files in your browser, so you do not have to convert first and the video itself is never uploaded. If you only need the sound from a video, video to audio does that in your browser.
Good for
- Interview and podcast transcripts
- Captions and subtitles (SRT) for YouTube, Reels and TikTok
- Lecture and meeting notes
- Turning voice memos into drafts
Honest limits
- Free: 10 minutes a day. Paid plans: 5 or 20 hours a month.
- Files up to 25 MB. For bigger videos, pull the audio out first with video to audio and drop the MP3.
- Accuracy drops with background music, cross-talk and poor microphones. Edit before you publish.
- No speaker labels, no translation, no live microphone transcription.
The other direction
Have a transcript and need a voice? The text to audio converter turns it back into speech, and the voice over generator adds timing.
Questions
Is the audio to text converter free?
Yes: 10 minutes of audio a day, no sign-up, no watermark on exports. Creator includes 5 hours a month and Studio 20 hours.
Which files can I transcribe?
Anything your browser can play: MP3, WAV, M4A, AAC, OGG, FLAC and WebM audio, plus MP4, MOV and WebM video. Files up to 25 MB. Your browser pulls out the audio track and shrinks it to a small mono file before anything is uploaded, so a video never leaves your device whole.
How accurate is the transcription?
On clear speech with one or two speakers, expect to fix a few words per minute, mostly names and jargon. Music under speech, cross-talk and phone microphones lower accuracy, and timestamps can drift by a second or so on long files. Every segment is editable before you export.
Can I get subtitles, not just text?
Yes. SRT and VTT keep the start and end time of each segment, ready for YouTube, Premiere, DaVinci Resolve or CapCut. TXT is plain paragraphs for notes and articles.
Which languages are supported?
The speech model detects the language on its own and handles English, Spanish, French, German, Portuguese, Italian, Dutch, Japanese, Korean, Chinese, Hindi, Arabic, Russian and many more. If detection picks the wrong one, choose the language before you drop the file.
What happens to my recordings?
The compressed audio is sent to our speech provider (Google Gemini, accessed through Kie.ai) to be transcribed and is deleted from the provider’s temporary storage within 24 hours. We do not keep the audio or the transcript on our servers. Details in the privacy policy.
Does it label speakers?
No. Transcripts are one stream of timed segments. Speaker labels are not offered.