Capture and playback failures were only written to the log on detached threads, so a microphone or output that could not be opened still showed a connected voice room with no sound. Threads now report into a shared diagnostics sink that the UI polls: errors appear in the status bar, a room with neither working half is marked failed, and signal counters (captured, sent, received, decoded, input peak) are logged every ten seconds. A microphone that delivers only digital silence is reported, and rejected voice datagrams are logged instead of dropped silently. Voice activation now tracks the background noise floor and opens just above it, capped by a much lower default ceiling; the old fixed threshold (about -26 dBFS) rarely opened for normal speech. Stream configuration only accepts f32 formats and says so when a device offers nothing else. Playback gains a mixing queue for short notification effects. |
||
|---|---|---|
| .cargo | ||
| docs | ||
| examples/themes | ||
| patches | ||
| protocol | ||
| scripts | ||
| src | ||
| tests | ||
| translations | ||
| .gitattributes | ||
| .gitignore | ||
| build.rs | ||
| Cargo.lock | ||
| Cargo.toml | ||
| CHANGELOG.md | ||
| LICENSE | ||
| README.md | ||
VNOX — Client
Native desktop client for VNOX. Voice, text, no cloud.
What is this?
The official desktop client for VNOX — a self-hosted voice and text communication platform. Connects to a VNOX server over the LNEx protocol.
- Connects to gateway via TCP (text, auth, channels, guilds, DMs)
- Connects to voice node via UDP (encrypted Opus voice packets)
- Opus audio encoding/decoding with adaptive jitter buffer
- Native UI built with Slint — no Electron, no web views
Features
Implemented
- Auth: Ed25519 keypair generation, challenge-response login, reconnect with exponential backoff
- Encryption: TLS 1.3 (rustls) on the control connection with certificate pinning (trust on first use), ChaCha20-Poly1305 on UDP voice
- Text chat: Persistent history, reactions, replies, edit, delete, typing indicators, read receipts
- Channel management: Join, leave, create, delete; text and voice channels
- Guilds: Create, list, leave, member list with roles, kick, audit log viewer
- Roles: Role list, assign/unassign with permission checks
- Direct Messages: 1:1 DMs with persistent history, unread badges, search
- Friends: Requests, accept/decline, Online/All/Pending/Blocked tabs
- Presence: Online/Idle/DND/Invisible, custom status text, activity display
- Voice: PTT / VAD / always-on modes, configurable bitrate, per-user volume
- Jitter buffer: Adaptive mode, configurable target latency
- Noise suppression: RNNoise (feature-gated, off by default)
- Identity vault: Optional Argon2id + ChaCha20-Poly1305 encrypted keyfile
- Keyfile export/import: Encrypted or plain JSON with passphrase
- Bookmarks: Save/remove server nodes in connect screen
Planned
- In-game overlay (Phase 2)
- Federation support (Phase 3)
- Mobile client (Phase 3)
Status
Phase 1 is implemented. Not production ready.
| Feature | Status |
|---|---|
| Connect to server | ✅ Working |
| Text chat | ✅ Working |
| Channel management | ✅ Working |
| Guilds & roles | ✅ Working |
| Direct Messages | ✅ Working |
| Friends & presence | ✅ Working |
| Voice (send/receive) | 🔧 Partial (audio pipeline) |
| Encryption | ✅ TLS 1.3 (TCP) + ChaCha20-Poly1305 (voice) |
| In-game overlay | 🔲 Phase 2 |
| Mobile client | 🔲 Phase 3 |
See docs/00-status.md for a full breakdown.
Quick start
Requirements: Rust 1.94+, a running VNOX Server
Linux builds also need audio/protocol prerequisites and FreeType/fontconfig development libraries (libfreetype6-dev libfontconfig1-dev on Ubuntu); native file selection uses the desktop portal.
# Terminal 1 — gateway
cargo run -p vnox-gateway -- --config dev/config.toml
# Terminal 2 — voice node
cargo run -p vnox-voice-node -- --config dev/config.toml
# Terminal 3 — client
cargo run -p vnox-client
The client connects to 127.0.0.1:7600 by default. You can change the address in the UI on the connect screen.
Opus on Windows
audiopus_sys builds libopus from source via CMake. CMake 4.x policy flag is already set in .cargo/config.toml — no manual steps needed.
Tech stack
| Layer | Technology |
|---|---|
| UI | Slint |
| Networking | Tokio (async TCP + UDP) |
| Audio | cpal + rodio + opus + rnnoise |
| Identity | ed25519-dalek + x25519-dalek |
| Crypto | ChaCha20-Poly1305 + Argon2id |
| Protocol | LNEx v1 (JSON, Phase 1) |
Project structure
src/
├── main.rs # Entry point
├── ui/ # Slint UI (slint files + Rust glue)
├── net/ # LNEx TCP + UDP networking
│ ├── crypto/ # Session encryption
│ ├── framing/ # Packet read/write
│ ├── voice/ # UDP voice send/recv
│ └── session/ # Connection lifecycle, reconnection
├── audio/ # Audio pipeline
│ ├── capture.rs # Mic → Opus encode
│ ├── playback.rs # Opus decode → speaker
│ ├── config.rs # Bitrate, VAD, jitter settings
│ └── processing/ # VAD, noise suppression
├── identity_vault.rs # Argon2id-encrypted keyfile
├── app.rs # UI state and event dispatch
└── jitter/ # Jitter buffer
License
Client code: GPL-3.0 — see LICENSE
LNEx protocol specification: CC0 (public domain)