wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

format with ruff

Browse Source
This commit is contained in:
joachimchauvet
2024-09-24 10:21:02 +03:00
parent fa609f1afc
commit b6e1d6e6ae
2 changed files with 69 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
examples/foundational/01b-livekit-audio.py
View File

@@ -35,7 +35,9 @@ def generate_token(room_name: str, participant_name: str, api_key: str, api_secr
async def configure_livekit(): async def configure_livekit():
parser = argparse.ArgumentParser(description="LiveKit AI SDK Bot Sample") parser = argparse.ArgumentParser(description="LiveKit AI SDK Bot Sample")
parser.add_argument("-r", "--room", type=str, required=False, help="Name of the LiveKit room to join") parser.add_argument(
"-r", "--room", type=str, required=False, help="Name of the LiveKit room to join"
)
parser.add_argument("-u", "--url", type=str, required=False, help="URL of the LiveKit server") parser.add_argument("-u", "--url", type=str, required=False, help="URL of the LiveKit server")
args, unknown = parser.parse_known_args() args, unknown = parser.parse_known_args()
@@ -56,7 +58,9 @@ async def configure_livekit():
) )
if not api_key or not api_secret: if not api_key or not api_secret:
raise Exception("LIVEKIT_API_KEY and LIVEKIT_API_SECRET must be set in environment variables.") raise Exception(
"LIVEKIT_API_KEY and LIVEKIT_API_SECRET must be set in environment variables."
)
token = generate_token(room_name, "Say One Thing", api_key, api_secret) token = generate_token(room_name, "Say One Thing", api_key, api_secret)
@@ -71,7 +75,10 @@ async def main():
(url, token, room_name) = await configure_livekit() (url, token, room_name) = await configure_livekit()
transport = LiveKitTransport( transport = LiveKitTransport(
url=url, token=token, room_name=room_name, params=LiveKitParams(audio_out_enabled=True, audio_out_sample_rate=16000) url=url,
token=token,
room_name=room_name,
params=LiveKitParams(audio_out_enabled=True, audio_out_sample_rate=16000),
) )
tts = CartesiaTTSService( tts = CartesiaTTSService(
@@ -88,7 +95,11 @@ async def main():
@transport.event_handler("on_first_participant_joined") @transport.event_handler("on_first_participant_joined")
async def on_first_participant_joined(transport, participant_id): async def on_first_participant_joined(transport, participant_id):
await asyncio.sleep(1) await asyncio.sleep(1)
await task.queue_frame(TextFrame("Hello there! How are you doing today? Would you like to talk about the weather?")) await task.queue_frame(
TextFrame(
"Hello there! How are you doing today? Would you like to talk about the weather?"
)
)
await runner.run(task) await runner.run(task)

70
src/pipecat/transports/services/livekit.py
View File

@@ -109,8 +109,12 @@ class LiveKitTransportClient:
logger.info(f"Connected to {self._room_name}") logger.info(f"Connected to {self._room_name}")
# Set up audio source and track # Set up audio source and track
self._audio_source = rtc.AudioSource(self._params.audio_out_sample_rate, self._params.audio_out_channels) self._audio_source = rtc.AudioSource(
self._audio_track = rtc.LocalAudioTrack.create_audio_track("pipecat-audio", self._audio_source) self._params.audio_out_sample_rate, self._params.audio_out_channels
)
self._audio_track = rtc.LocalAudioTrack.create_audio_track(
"pipecat-audio", self._audio_source
)
options = rtc.TrackPublishOptions() options = rtc.TrackPublishOptions()
options.source = rtc.TrackSource.SOURCE_MICROPHONE options.source = rtc.TrackSource.SOURCE_MICROPHONE
await self._room.local_participant.publish_track(self._audio_track, options) await self._room.local_participant.publish_track(self._audio_track, options)
@@ -136,7 +140,9 @@ class LiveKitTransportClient:
try: try:
if participant_id: if participant_id:
await self._room.local_participant.publish_data(data, reliable=True, destination_identities=[participant_id]) await self._room.local_participant.publish_data(
data, reliable=True, destination_identities=[participant_id]
)
else: else:
await self._room.local_participant.publish_data(data, reliable=True) await self._room.local_participant.publish_data(data, reliable=True)
except Exception as e: except Exception as e:
@@ -190,12 +196,18 @@ class LiveKitTransportClient:
asyncio.create_task(self._async_on_participant_disconnected(participant)) asyncio.create_task(self._async_on_participant_disconnected(participant))
def _on_track_subscribed_wrapper( def _on_track_subscribed_wrapper(
self, track: rtc.Track, publication: rtc.RemoteTrackPublication, participant: rtc.RemoteParticipant self,
track: rtc.Track,
publication: rtc.RemoteTrackPublication,
participant: rtc.RemoteParticipant,
): ):
asyncio.create_task(self._async_on_track_subscribed(track, publication, participant)) asyncio.create_task(self._async_on_track_subscribed(track, publication, participant))
def _on_track_unsubscribed_wrapper( def _on_track_unsubscribed_wrapper(
self, track: rtc.Track, publication: rtc.RemoteTrackPublication, participant: rtc.RemoteParticipant self,
track: rtc.Track,
publication: rtc.RemoteTrackPublication,
participant: rtc.RemoteParticipant,
): ):
asyncio.create_task(self._async_on_track_unsubscribed(track, publication, participant)) asyncio.create_task(self._async_on_track_unsubscribed(track, publication, participant))
@@ -218,7 +230,10 @@ class LiveKitTransportClient:
await self._callbacks.on_participant_disconnected(participant.sid) await self._callbacks.on_participant_disconnected(participant.sid)
async def _async_on_track_subscribed( async def _async_on_track_subscribed(
self, track: rtc.Track, publication: rtc.RemoteTrackPublication, participant: rtc.RemoteParticipant self,
track: rtc.Track,
publication: rtc.RemoteTrackPublication,
participant: rtc.RemoteParticipant,
): ):
if track.kind == rtc.TrackKind.KIND_AUDIO: if track.kind == rtc.TrackKind.KIND_AUDIO:
logger.info(f"Audio track subscribed: {track.sid} from participant {participant.sid}") logger.info(f"Audio track subscribed: {track.sid} from participant {participant.sid}")
@@ -227,7 +242,10 @@ class LiveKitTransportClient:
asyncio.create_task(self._process_audio_stream(audio_stream, participant.sid)) asyncio.create_task(self._process_audio_stream(audio_stream, participant.sid))
async def _async_on_track_unsubscribed( async def _async_on_track_unsubscribed(
self, track: rtc.Track, publication: rtc.RemoteTrackPublication, participant: rtc.RemoteParticipant self,
track: rtc.Track,
publication: rtc.RemoteTrackPublication,
participant: rtc.RemoteParticipant,
): ):
logger.info(f"Track unsubscribed: {publication.sid} from {participant.identity}") logger.info(f"Track unsubscribed: {publication.sid} from {participant.identity}")
if track.kind == rtc.TrackKind.KIND_AUDIO: if track.kind == rtc.TrackKind.KIND_AUDIO:
@@ -268,7 +286,9 @@ class LiveKitInputTransport(BaseInputTransport):
self._vad_analyzer: VADAnalyzer | None = params.vad_analyzer self._vad_analyzer: VADAnalyzer | None = params.vad_analyzer
self._current_sample_rate: int = params.audio_in_sample_rate self._current_sample_rate: int = params.audio_in_sample_rate
if params.vad_enabled and not params.vad_analyzer: if params.vad_enabled and not params.vad_analyzer:
self._vad_analyzer = VADAnalyzer(sample_rate=self._current_sample_rate, num_channels=self._params.audio_in_channels) self._vad_analyzer = VADAnalyzer(
sample_rate=self._current_sample_rate, num_channels=self._params.audio_in_channels
)
async def start(self, frame: StartFrame): async def start(self, frame: StartFrame):
await super().start(frame) await super().start(frame)
@@ -326,7 +346,9 @@ class LiveKitInputTransport(BaseInputTransport):
except Exception as e: except Exception as e:
logger.error(f"Error in audio input task: {e}") logger.error(f"Error in audio input task: {e}")
def _convert_livekit_audio_to_pipecat(self, audio_frame_event: rtc.AudioFrameEvent) -> AudioRawFrame: def _convert_livekit_audio_to_pipecat(
self, audio_frame_event: rtc.AudioFrameEvent
) -> AudioRawFrame:
audio_frame = audio_frame_event.frame audio_frame = audio_frame_event.frame
audio_data = np.frombuffer(audio_frame.data, dtype=np.int16) audio_data = np.frombuffer(audio_frame.data, dtype=np.int16)
original_sample_rate = audio_frame.sample_rate original_sample_rate = audio_frame.sample_rate
@@ -340,11 +362,19 @@ class LiveKitInputTransport(BaseInputTransport):
if sample_rate != self._current_sample_rate: if sample_rate != self._current_sample_rate:
self._current_sample_rate = sample_rate self._current_sample_rate = sample_rate
self._vad_analyzer = VADAnalyzer(sample_rate=self._current_sample_rate, num_channels=self._params.audio_in_channels) self._vad_analyzer = VADAnalyzer(
sample_rate=self._current_sample_rate, num_channels=self._params.audio_in_channels
)
return AudioRawFrame(audio=audio_data.tobytes(), sample_rate=sample_rate, num_channels=audio_frame.num_channels) return AudioRawFrame(
audio=audio_data.tobytes(),
sample_rate=sample_rate,
num_channels=audio_frame.num_channels,
)
def _resample_audio(self, audio_data: np.ndarray, original_rate: int, target_rate: int) -> np.ndarray: def _resample_audio(
self, audio_data: np.ndarray, original_rate: int, target_rate: int
) -> np.ndarray:
num_samples = int(len(audio_data) * target_rate / original_rate) num_samples = int(len(audio_data) * target_rate / original_rate)
resampled_audio = signal.resample(audio_data, num_samples) resampled_audio = signal.resample(audio_data, num_samples)
return resampled_audio.astype(np.int16) return resampled_audio.astype(np.int16)
@@ -392,7 +422,9 @@ class LiveKitOutputTransport(BaseOutputTransport):
if hasattr(frame, "characters"): if hasattr(frame, "characters"):
metrics["characters"] = frame.characters metrics["characters"] = frame.characters
message = LiveKitTransportMessageFrame(message={"type": "pipecat-metrics", "metrics": metrics}) message = LiveKitTransportMessageFrame(
message={"type": "pipecat-metrics", "metrics": metrics}
)
await self._client.send_data(str(message.message).encode()) await self._client.send_data(str(message.message).encode())
async def write_raw_audio_frames(self, frames: bytes): async def write_raw_audio_frames(self, frames: bytes):
@@ -430,7 +462,9 @@ class LiveKitTransport(BaseTransport):
self._room_name = room_name self._room_name = room_name
self._params = params self._params = params
self._client = LiveKitTransportClient(url, token, room_name, self._params, self._create_callbacks(), self._loop) self._client = LiveKitTransportClient(
url, token, room_name, self._params, self._create_callbacks(), self._loop
)
self._input: LiveKitInputTransport | None = None self._input: LiveKitInputTransport | None = None
self._output: LiveKitOutputTransport | None = None self._output: LiveKitOutputTransport | None = None
@@ -463,7 +497,9 @@ class LiveKitTransport(BaseTransport):
def output(self) -> FrameProcessor: def output(self) -> FrameProcessor:
if not self._output: if not self._output:
self._output = LiveKitOutputTransport(self._client, self._params, name=self._output_name) self._output = LiveKitOutputTransport(
self._client, self._params, name=self._output_name
)
return self._output return self._output
@property @property
@@ -519,7 +555,9 @@ class LiveKitTransport(BaseTransport):
participant = self._client._room.remote_participants.get(participant_id) participant = self._client._room.remote_participants.get(participant_id)
if participant: if participant:
for publication in participant.audio_tracks.values(): for publication in participant.audio_tracks.values():
self._client._on_track_subscribed_wrapper(publication.track, publication, participant) self._client._on_track_subscribed_wrapper(
publication.track, publication, participant
)
async def _on_audio_track_unsubscribed(self, participant_id: str): async def _on_audio_track_unsubscribed(self, participant_id: str):
await self._call_event_handler("on_audio_track_unsubscribed", participant_id) await self._call_event_handler("on_audio_track_unsubscribed", participant_id)