Back to the catalogue
SpeechIngestion
Whisper Large v3
Speech-to-text for voice notes in chat and for indexing recorded calls and meetings.
Two uses, one model. In chat, press the microphone and the recording is transcribed before it becomes your question — the audio travels with the turn, so you can play back what was actually said.
In a flow, a speech-to-text step turns a call recording into text a later step can search, categorise or summarise. Billing is by audio duration, and a recording whose length the provider does not report is refused rather than charged at a guess.