SignalForge Audio

Planned

The SignalForge Audio subsystem handles audio capture, processing, and playback.

Features

• Low-latency audio pipeline (target: <10ms round-trip)
• Sample-accurate mixing for multiple audio sources
• Hardware abstraction through NPL audio drivers
• Capability-controlled microphone access
• Integration with the Surface Manager for audio-visual synchronization

Pipeline Model

Microphone (NPL) → Capture → Processing → Mix → Playback (NPL)
                         ↓                    ↑
                    App ION Ring          App ION Ring

Applications connect to the audio pipeline through ION Rings. Multiple applications can submit audio simultaneously — the mixer combines them with sample-accurate timing.

Security

Microphone access requires an explicit capability grant (AUDIO_CAPTURE). The grant is:

• Logged to ProvChain
• Visible in the HUD (a recording indicator)
• Revocable by the user at any time
• Subject to pledge restrictions

An application cannot capture audio without the user's knowledge. The capability system makes covert audio recording architecturally impossible.