Talon
186fe6dbb6
The screen-audio pump opened and disposed a memory mapping plus its container lookup and path strings on every 5 ms tick - about 200 mapping pairs per second of steady allocation that churned the GC under long calls. The producer opens the ring with O_CREAT and never replaces it, so the pump now keeps one mapping across ticks and rebuilds it only when the ring file disappears or a drain fails.
VoiceCat
VoiceCat is a self-hosted, channel-based voice and text chat system built on .NET 10. It uses TLS 1.3 for protobuf control traffic and authenticated encrypted UDP for Opus media. There is no WebRTC, central directory, or plaintext mode.
The repository contains a managed server, CLI, shared client/audio core, and native Windows, macOS, and iOS user interfaces. A small C library supplies Opus/RNNoise, and a small Swift iOS extension captures ReplayKit application audio.
Build
./scripts/build-native.ps1
dotnet restore VoiceCat.slnx --locked-mode
dotnet build VoiceCat.slnx -c Release --no-restore
dotnet test VoiceCat.slnx -c Release --no-build
See CLAUDE.md for the developer map, docs/README.md for current contracts, and PROGRESS.md for the short release handoff.
Layout
proto/ protobuf wire schema
src/ managed protocol, crypto, server, client, audio, and CLI
tests/ managed behavior and integration tests
clients/windows/ WinForms client
clients/apple/ AppKit and UIKit clients
native/media/ narrow Opus/RNNoise C shim
native/rnnoise/ vendored RNNoise source and model
native/apple/broadcast/ ReplayKit broadcast extension
docs/ current contracts and operating documentation
Non-negotiable constraints
- Encryption is mandatory.
- No GPL or LGPL dependencies.
- Real-time audio callbacks never allocate, lock, block, or perform I/O.
- Accessibility is a release requirement on every client platform.