What is it?

Whisper is the transcription engine Performy uses to convert every meeting's audio into text. It's different from the AI providers you connect with your own account (OpenAI, Anthropic, Google Gemini): those analyze the conversation once it's already transcribed, while Whisper is the step before that — it turns speech into text so the analysis can run on it.

How it works: it's automatic. No account to connect, no API key to set up — it runs on Performy's own infrastructure over every meeting you upload.
1

Upload or analyze a meeting

As soon as the meeting's audio or video finishes processing, Performy sends it to Whisper for transcription — no extra step on your end.

2

The transcript is generated

Whisper returns the full text of the conversation, which Performy stores alongside the meeting.

3

View it in the meeting detail

The Transcription tab in the meeting detail page shows the full transcript and lets you download it, along with the original recording.

4

It powers the analysis

The text Whisper transcribes is what your AI provider (OpenAI, Anthropic, or Google Gemini) then analyzes to evaluate performance and playbook adherence.