Use of AI
Riff Repeat uses two AI models to power its transcription and summary features. This page explains what each model does and how your data is handled.
Whisper — Audio Transcription
Riff Repeat uses OpenAI Whisper to convert lesson audio into text transcripts. Whisper is a speech-to-text model — it does nothing beyond converting audio to words.
We run Whisper on privately configured compute infrastructure. Your audio is processed in an isolated environment and is not shared with any third party. No audio data leaves our controlled infrastructure during transcription.
Anthropic Claude — Summaries & Exercises
We use Anthropic's Claude to generate lesson summaries and exercises from transcripts. This means transcript text is sent to Anthropic's servers for processing.
However, Anthropic's API Terms of Service — which govern how developers use Claude, as distinct from consumer chat products — explicitly guarantee that data submitted via the API is not used to train future models. You can read more about this in Anthropic's own documentation. Your lesson content is used only to produce the output you requested, and nothing more.
- Audio (Whisper): Processed on private infrastructure — never leaves our control
- Transcripts (Claude): Sent to Anthropic's API, then discarded — not used for training
- We will NEVER use your data to train AI models
Questions?
If you have questions about how AI is used in Riff Repeat, please contact us or review our Privacy Policy.