sirmrak/linux-voice-assistant
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add LED Events to __main__.py

Browse Source
This commit is contained in:
imonlinux
2025-10-10 19:04:08 -07:00
committed by GitHub
parent bc8e6e0ef7
commit ec51b0b11e
1 changed files with 11 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files
linux_voice_assistant/__main__.py
+11 -33
View File
@@ -13,6 +13,8 @@ import numpy as np
import sounddevice as sd
from .microwakeword import MicroWakeWord, MicroWakeWordFeatures
from .event_bus import EventBus
from .event_led import LedEvent
from .models import AvailableWakeWord, Preferences, ServerState, WakeWordType
from .mpv_player import MpvMediaPlayer
from .openwakeword import OpenWakeWord, OpenWakeWordFeatures
@@ -62,7 +64,6 @@ async def main() -> None:
type=float,
help="Seconds before wake word can be activated again",
)
#
parser.add_argument(
"--oww-melspectrogram-model",
default=_OWW_DIR / "melspectrogram.tflite",
@@ -73,16 +74,13 @@ async def main() -> None:
default=_OWW_DIR / "embedding_model.tflite",
help="Path to openWakeWord embedding model",
)
#
parser.add_argument(
"--wakeup-sound", default=str(_SOUNDS_DIR / "wake_word_triggered.flac")
)
parser.add_argument(
"--timer-finished-sound", default=str(_SOUNDS_DIR / "timer_finished.flac")
)
#
parser.add_argument("--preferences-file", default=_REPO_DIR / "preferences.json")
#
parser.add_argument(
"--host",
default="0.0.0.0",
@@ -107,7 +105,6 @@ async def main() -> None:
for model_config_path in wake_word_dir.glob("*.json"):
model_id = model_config_path.stem
if model_id == args.stop_model:
# Don't show stop model as an available wake word
continue
with open(model_config_path, "r", encoding="utf-8") as model_config_file:
@@ -139,7 +136,6 @@ async def main() -> None:
# Load wake/stop models
wake_models: Dict[str, Union[MicroWakeWord, OpenWakeWord]] = {}
if preferences.active_wake_words:
# Load preferred models
for wake_word_id in preferences.active_wake_words:
wake_word = available_wake_words.get(wake_word_id)
if wake_word is None:
@@ -150,14 +146,11 @@ async def main() -> None:
wake_models[wake_word_id] = wake_word.load(libtensorflowlite_c_path)
if not wake_models:
# Load default model
wake_word_id = args.wake_model
wake_word = available_wake_words[wake_word_id]
_LOGGER.debug("Loading wake model: %s", wake_word_id)
wake_models[wake_word_id] = wake_word.load(libtensorflowlite_c_path)
# TODO: allow openWakeWord for "stop"
stop_model: Optional[MicroWakeWord] = None
for wake_word_dir in wake_word_dirs:
stop_config_path = wake_word_dir / f"{args.stop_model}.json"
@@ -192,6 +185,11 @@ async def main() -> None:
refractory_seconds=args.refractory_seconds,
)
# --- LED EVENT SYSTEM (from LVA) ---
state.event_bus = EventBus()
LedEvent(state)
state.event_bus.publish("ready", {})
process_audio_thread = threading.Thread(
target=process_audio, args=(state,), daemon=True
)
@@ -205,7 +203,6 @@ async def main() -> None:
lambda: VoiceSatelliteProtocol(state), host=args.host, port=args.port
)
# Auto discovery (zeroconf, mDNS)
discovery = HomeAssistantZeroconf(port=args.port, name=args.name)
await discovery.register_server()
@@ -231,9 +228,6 @@ async def main() -> None:
_LOGGER.debug("Server stopped")
# -----------------------------------------------------------------------------
def process_audio(state: ServerState):
"""Process audio chunks from the microphone."""
@@ -244,7 +238,6 @@ def process_audio(state: ServerState):
oww_features: Optional[OpenWakeWordFeatures] = None
oww_inputs: List[np.ndarray] = []
has_oww = False
last_active: Optional[float] = None
try:
@@ -252,25 +245,18 @@ def process_audio(state: ServerState):
audio_chunk = state.audio_queue.get()
if audio_chunk is None:
break
if state.satellite is None:
continue
if (not wake_words) or (state.wake_words_changed and state.wake_words):
# Update list of wake word models to process
state.wake_words_changed = False
wake_words = [ww for ww in state.wake_words.values() if ww.is_active]
has_oww = False
for wake_word in wake_words:
if isinstance(wake_word, OpenWakeWord):
has_oww = True
has_oww = any(isinstance(ww, OpenWakeWord) for ww in wake_words)
if micro_features is None:
micro_features = MicroWakeWordFeatures(
libtensorflowlite_c_path=state.libtensorflowlite_c_path,
)
if has_oww and (oww_features is None):
oww_features = OpenWakeWordFeatures(
melspectrogram_model=state.oww_melspectrogram_path,
@@ -280,7 +266,6 @@ def process_audio(state: ServerState):
try:
state.satellite.handle_audio(audio_chunk)
assert micro_features is not None
micro_inputs.clear()
micro_inputs.extend(micro_features.process_streaming(audio_chunk))
@@ -303,7 +288,6 @@ def process_audio(state: ServerState):
activated = True
if activated:
# Check refractory
now = time.monotonic()
if (last_active is None) or (
(now - last_active) > state.refractory_seconds
@@ -311,22 +295,16 @@ def process_audio(state: ServerState):
state.satellite.wakeup(wake_word)
last_active = now
# Always process to keep state correct
stopped = False
for micro_input in micro_inputs:
if state.stop_word.process_streaming(micro_input):
stopped = True
stopped = any(
state.stop_word.process_streaming(mi) for mi in micro_inputs
)
if stopped and state.stop_word.is_active:
state.satellite.stop()
except Exception:
_LOGGER.exception("Unexpected error handling audio")
except Exception:
_LOGGER.exception("Unexpected error processing audio")
# -----------------------------------------------------------------------------
if __name__ == "__main__":
asyncio.run(main())