Private and offline
WhisperX runs on your computer, also on the GPU of a Mac: free, unlimited and without sending your audio anywhere. The Whisper API and the Google API are available too.
Audiotext
Transcribe audio and video in 100 languages, on your computer and for free. Then play, correct, translate, summarize and subtitle the result.
Any source
Audio and video files, YouTube videos, links to media files, the microphone (with live text), whole folders, or a folder that is watched for new files.
Play and correct
Click a segment to play it, find and replace, rename the speakers, review the words it’s least sure of and fix them while keeping the timestamps.
Translate, summarize and ask
Get a summary, its key points and chapters, answers to your questions, or translations, with OpenAI, Claude, Gemini, DeepSeek, Mistral, Grok, Ollama, DeepL or Google Translate.
Subtitles and exports
Export to TXT, Markdown, Word, SRT, VTT, ASS, TSV or JSON, watch videos with their subtitles, or the ones of their translations, and burn them into the video.
In your language
The interface is available in 22 languages, and it can be used from the command line too.