Speech to text
Transcribe the speech in an audio or video file into text
Turns spoken audio at a web address into text with timed segments, using the open Whisper large-v3-turbo speech model.
What it asks for
- audio_url (Needed, web address): A web address that serves the audio file itself (mp3, wav, m4a, ogg, flac or webm), up to 25 MB.
Use it from your AI
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to transcribe the speech in an audio or video file into text”
What your AI calls: run_tool {"tool_id":"tool:speech-to-text"}