Speech to Text
Upload a recording or a clip and get the text back, with subtitle files
Drop an audio or video file here
mp3, m4a, wav, mp4, mov and more · up to 40 MB · up to 60 minutes
About this tool
Speech to text turns a recording or a video into text you can copy straight out and use. It suits meeting notes, summarising a lecture, writing up an interview, or subtitling your own clips — without replaying the audio and typing it out a sentence at a time.
It takes the formats people actually have: mp3, m4a (what an iPhone records), wav, opus, amr, and video files such as mp4, mov and mkv, from which only the audio track is used. Results download in three shapes — TXT to carry on writing with, and SRT and VTT for subtitles in an editor or on YouTube.
What sets it apart from the usual transcription services is that your file never leaves our machine. Nothing is sent to anyone's API. The audio is deleted as soon as the transcript exists and the text goes within 30 minutes — which matters a great deal when what you are transcribing is an internal meeting, an interview, or a medical note.
Worth knowing before you start: files are limited to 40MB and 60 minutes, overlapping speakers and noisy rooms transcribe less accurately than clear speech, and the result should always be read through before you rely on it — especially numbers, names and technical terms, which is where every transcription system makes its mistakes.
How to use
- 1Drop an audio or video file in, or click to choose one (up to 40MB, up to 60 minutes)
- 2Pick the language spoken in the file — Thai is selected for you
- 3Press “Start transcribing” and leave the page open; progress is shown as a percentage
- 4Copy the text, or download it as TXT / SRT / VTT for subtitles
Frequently asked questions
- How accurate is it for Thai?
- It runs the full Whisper large-v3 model, which is the most accurate one for Thai — not the smaller, faster variant, whose accuracy drop is worst in Thai specifically. Clear speech in a quiet room comes out very well; background music, several people talking at once, or a distant microphone will produce more mistakes, as they do for every system.
- Is my audio stored anywhere?
- No. The audio is deleted the moment the transcript is ready, and the text itself is deleted within 30 minutes. We keep no copy, never log the filename, and send nothing to any third-party service — it is all processed on our own machine.
- Can I use this to subtitle a video?
- Yes. Download the SRT or VTT file and load it into YouTube, Premiere Pro, CapCut or any other editor — each line already carries its start and end time.
- Can I upload a video directly, or do I need to extract the audio first?
- Upload the video directly (mp4, mov, mkv and others). Only the audio track is used; the picture is discarded immediately and never processed.
- Why does it sometimes say it is temporarily unavailable?
- Transcription needs a graphics card, which lives on our own machine. If that machine is off or being updated, this tool says so plainly rather than letting you upload and wait for nothing. No other tool on the site is affected.
- How long does it take?
- Roughly a third of the recording's length — a 15-minute file takes about 5 minutes. If other people are using it you join a queue and your position is shown.