How do I transcribe audio to text?
Updated:
Open AI Audio to Text, drop the recording and press Transcribe. Whisper turns the speech into text you can copy, or save as a .txt file or .srt subtitles. It is part of AudioCutter Pro and runs privately in your browser or on a fast server.
What are the steps?
- Open Audio to Text. It is part of Pro, with a 7-day pass or a 1-year license.
- Choose Fast (server) for speed and accuracy, or Private (in this browser) to keep the file on your device.
- Pick the language of the audio, or leave Detect automatically on.
- Drop an audio or video file and press Transcribe.
- Copy the transcript, or download it as .txt or .srt subtitles.
How accurate is the transcription?
Fast mode uses a large Whisper model and handles accents, background noise and fast speakers well. Private mode is slower and less accurate, especially on phones, but the audio never leaves your device.
How long can the recording be?
Fast mode takes up to 1 hour per file and splits long files automatically. Private mode transcribes the first 10 minutes of a file.
Can I make subtitles for a video?
Yes. Drop the video itself, then download the .srt file: it holds the text with timings, and YouTube, Premiere, CapCut and most players accept it.
Which languages can it transcribe?
Around a hundred with Detect automatically, including English, Spanish, Hindi, Arabic, Portuguese, French and Japanese. You can also pick one of 13 languages yourself, which helps with short clips.
Where does my audio go?
In Private mode it stays in your browser. In Fast mode it is sent to DeepInfra in the US for transcription and is not stored.