by r-dh
Transcribe audio and video files with speaker detection, timestamps, and multiple export formats. Powered by Mistral's Voxtral model.
Paste your Mistral API key to get started. Get one here.
Your key is stored locally in this browser and never sent anywhere except Mistral.
Choose a file or drag it here
Audio or video, up to 1 GB
Tip: if names come out wrong, add them in Settings next time.
Stored locally in this browser
Names, brands, or words the transcription might misspell.
Each term must be a single word. Separate multiple words with commas.
Identify who said what