162 lines
6.7 KiB
TOML
162 lines
6.7 KiB
TOML
[package]
|
|
name = "chanora_audio"
|
|
description = "Chanora audio subsystem — platform-native capture/playback, audiopus encode, tsclientlib AudioHandler for decode + jitter buffer + mix. DEC-011, DEC-011.1."
|
|
version.workspace = true
|
|
edition.workspace = true
|
|
rust-version.workspace = true
|
|
authors.workspace = true
|
|
license.workspace = true
|
|
repository.workspace = true
|
|
publish.workspace = true
|
|
|
|
[features]
|
|
default = []
|
|
bench-harness = ["dep:criterion", "dep:dhat", "dep:serde_json"]
|
|
|
|
[dependencies]
|
|
chanora_protocol = { path = "../chanora_protocol" }
|
|
thiserror.workspace = true
|
|
tracing.workspace = true
|
|
sonora = "0.1"
|
|
webrtc-vad = "0.4"
|
|
|
|
# ndarray is required by ort's tensor construction API and by
|
|
# Silero VAD ONNX inference across all platforms.
|
|
ndarray = "0.17"
|
|
|
|
# Opus encoder. tsclientlib already pulls this; we depend explicitly so
|
|
# this crate can compile against it without going through tsclientlib.
|
|
audiopus = "0.3.0-rc.0"
|
|
# AudioHandler lives in the tsclientlib crate behind the `audio`
|
|
# feature. We import the crate just for the AudioHandler type; the
|
|
# Connection type stays inside chanora_protocol.
|
|
tsclientlib = { git = "https://github.com/ReSpeak/tsclientlib.git", rev = "04aa2491", default-features = false, features = ["audio"] }
|
|
tokio = { version = "1", features = ["sync", "rt", "macros", "time"] }
|
|
rustfft = "6.2.0"
|
|
criterion = { version = "0.5", optional = true }
|
|
dhat = { version = "0.3", optional = true }
|
|
serde_json = { version = "1", optional = true }
|
|
|
|
[target.'cfg(all(not(target_os = "android"), not(target_os = "ios"), not(target_os = "macos")))'.dependencies]
|
|
# Desktop audio I/O for Windows capture/playback and Linux fallback metadata.
|
|
# Linux voice I/O uses PipeWire first and PulseAudio as fallback; Apple platforms use direct VoiceProcessingIO
|
|
# AudioUnits via `coreaudio-rs` for the voice path.
|
|
cpal = "0.17.3"
|
|
|
|
[target.'cfg(any(target_os = "ios", target_os = "macos"))'.dependencies]
|
|
# Direct CoreAudio AudioUnit access on Apple platforms (DEC-011 follow-up).
|
|
# cpal's Apple path does not expose the voice-processing controls Chanora
|
|
# needs for VoIP. Use `kAudioUnitSubType_VoiceProcessingIO` directly via
|
|
# `coreaudio-rs` so capture/playback share Apple's native AEC + AGC + NS
|
|
# voice unit on both iOS and macOS.
|
|
#
|
|
# Default features keep `audio_toolbox` + `core_audio`, both required
|
|
# for AudioUnit construction + property access.
|
|
coreaudio-rs = "0.14"
|
|
# Grand Central Dispatch bindings — used to run AudioUnit initialize/start
|
|
# on the main queue to avoid the VPIO RPC timeout on iOS simulator.
|
|
dispatch2 = "0.3"
|
|
|
|
[target.'cfg(target_os = "ios")'.dependencies]
|
|
# ONNX Runtime Rust binding for Silero VAD v6 (P1 VAD_002). The official
|
|
# iOS CocoaPod ships ONNX Runtime as a static framework, so iOS links it
|
|
# into chanora_bridge at build time instead of loading a dylib at runtime.
|
|
ort = { version = "2.0.0-rc.12", default-features = false, features = ["std", "ndarray"] }
|
|
|
|
[target.'cfg(not(target_os = "ios"))'.dependencies]
|
|
ort = { version = "2.0.0-rc.12", default-features = false, features = ["load-dynamic", "ndarray", "api-24"] }
|
|
|
|
[target.'cfg(target_os = "android")'.dependencies]
|
|
# JNI bindings to flip Android's AudioManager into MODE_IN_COMMUNICATION
|
|
# when the voice-comm preset is requested. ndk_context is initialised
|
|
# by the bridge crate's android_init shim.
|
|
jni = { version = "0.21", default-features = false }
|
|
ndk-context = "0.1"
|
|
# Oboe-rs (Google Oboe wrapper) for low-latency voice capture + playback.
|
|
# Primary backend for SDD-111..SDD-115. The pre-compiled static library
|
|
# shipped with `oboe-sys` 0.6 covers armv7 / aarch64 / x86 / x86_64.
|
|
# Default features keep the precompiled library + pregenerated bindings
|
|
# so we avoid the clang-sys / libclang requirement on the build host.
|
|
#
|
|
# Using local fork edisonjwa/oboe-rs (v0.6.2) with:
|
|
# - catch_unwind safety in callbacks
|
|
# - unwrap_or_default in enum getters (no more SessionId panic)
|
|
# - get_raw_session_id() for JNI hardware effect binding
|
|
# - deduplicated macro impls
|
|
# - PowerSavingOffloaded PerformanceMode variant
|
|
oboe = { path = "../../../oboe-rs" }
|
|
|
|
[target.'cfg(target_os = "windows")'.dependencies]
|
|
# Real Windows global PTT (SDD-083 / SDD-084): RegisterRawInputDevices
|
|
# + WM_INPUT translation backed by a hidden message-only window, and
|
|
# SetWindowsHookExW(WH_KEYBOARD_LL / WH_MOUSE_LL) fallback. Both
|
|
# require a per-backend OS thread that owns a message pump.
|
|
windows = { version = "0.54", features = [
|
|
"Win32_Foundation",
|
|
"Win32_Graphics_Gdi",
|
|
"Win32_System_LibraryLoader",
|
|
"Win32_System_Threading",
|
|
"Win32_UI_Input",
|
|
"Win32_UI_Input_KeyboardAndMouse",
|
|
"Win32_UI_WindowsAndMessaging",
|
|
] }
|
|
|
|
[dev-dependencies]
|
|
# `test-util` enables `start_paused` / virtual-clock tests used by
|
|
# the missed-key-up watchdog unit tests.
|
|
tokio = { version = "1", features = ["sync", "rt", "macros", "time", "test-util"] }
|
|
# Cross-platform recording Layer for the SDD-090 / DEC-027 privacy
|
|
# invariant integration test (`tests/ptt_privacy.rs`).
|
|
tracing-subscriber = { version = "0.3", features = ["registry"] }
|
|
[[bench]]
|
|
name = "realtime_capture"
|
|
harness = false
|
|
path = "benches/realtime_capture.rs"
|
|
required-features = ["bench-harness"]
|
|
|
|
[[bench]]
|
|
name = "opus_codec"
|
|
harness = false
|
|
path = "benches/opus_codec.rs"
|
|
required-features = ["bench-harness"]
|
|
|
|
[[bench]]
|
|
name = "resampler"
|
|
harness = false
|
|
path = "benches/resampler.rs"
|
|
required-features = ["bench-harness"]
|
|
|
|
[[example]]
|
|
name = "compare_baseline"
|
|
path = "examples/compare_baseline.rs"
|
|
required-features = ["bench-harness"]
|
|
|
|
[[example]]
|
|
name = "emit_baseline"
|
|
path = "examples/emit_baseline.rs"
|
|
required-features = ["bench-harness"]
|
|
|
|
[target.'cfg(target_os = "linux")'.dependencies]
|
|
# GNOME-on-Wayland Global Push-to-Talk uses the freedesktop
|
|
# `org.freedesktop.portal.GlobalShortcuts` interface over D-Bus.
|
|
# `zbus` is the standard async D-Bus crate; the `tokio` runtime
|
|
# selector is mandatory in zbus 5; we share the tokio runtime
|
|
# the rest of the audio + core crates already depend on. The
|
|
# `blocking-api` feature is retained so the audio-engine
|
|
# probe path can do a synchronous portal-version read without
|
|
# starting an async runtime; the live session flow uses the
|
|
# async surface.
|
|
zbus = { version = "5", default-features = false, features = ["tokio", "blocking-api"] }
|
|
# Stream / sink utilities for consuming portal signals on the
|
|
# async path.
|
|
futures-util = { version = "0.3", default-features = false, features = ["std"] }
|
|
# Random token bytes for the portal handle_token / session_handle_token
|
|
# options. The portal recommends fresh tokens to scope its own
|
|
# object paths per call.
|
|
rand = "0.8"
|
|
# Linux-native voice I/O. PipeWire is primary on modern desktops;
|
|
# PulseAudio is retained as fallback and for older environments.
|
|
pipewire = { version = "0.10", features = ["v0_3_44"] }
|
|
libpulse-binding = "2.30"
|
|
libpulse-simple-binding = "2.29"
|