Local-first by design
Resonant processes all audio on your device. Your voice is never uploaded to a server for transcription. The speech-to-text models run locally using your device’s hardware:- Mac: Apple Neural Engine (ANE) for maximum performance
- Windows: CPU or GPU inference using optimized local models
What stays on your device
What leaves your device (opt-in only)
These features are optional and clearly marked in the app:Zero data retention
When transcript text is sent to cloud LLMs for cleanup or summarization, it is processed in real-time and not stored. Resonant uses zero-data-retention API agreements with all LLM providers.No telemetry on audio
Resonant collects basic app usage telemetry (feature usage, crash reports) via PostHog, but never collects or transmits audio data, transcript content, or meeting recordings.HIPAA Mode
For healthcare and high-security environments, Resonant offers a HIPAA-compatible configuration:- All cloud features disabled
- Local-only encryption for stored transcripts
- Audit logging for access events
- No telemetry of any kind