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

PR feedback

Browse Source
This commit is contained in:
Adrian Cowham
2024-10-11 11:29:01 -07:00
parent 917e482876
commit 083d221dd2
1 changed files with 5 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
src/pipecat/processors/audio/audio_buffer_processor.py
View File

@@ -14,7 +14,7 @@ from pipecat.processors.frame_processor import FrameDirection, FrameProcessor
class AudioBufferProcessor(FrameProcessor): class AudioBufferProcessor(FrameProcessor):
def __init__(self): def __init__(self, **kwargs):
""" """
Initialize the AudioBufferProcessor. Initialize the AudioBufferProcessor.
@@ -22,13 +22,11 @@ class AudioBufferProcessor(FrameProcessor):
- audio_buffer: A bytearray to store incoming audio data. - audio_buffer: A bytearray to store incoming audio data.
- num_channels: The number of audio channels (initialized as None). - num_channels: The number of audio channels (initialized as None).
- sample_rate: The sample rate of the audio (initialized as None). - sample_rate: The sample rate of the audio (initialized as None).
- assistant_audio: A boolean flag to indicate if assistant audio is being processed.
- user_audio: A boolean flag to indicate if user audio is being processed.
The num_channels and sample_rate are set to None initially and will be The num_channels and sample_rate are set to None initially and will be
populated when the first audio frame is processed. populated when the first audio frame is processed.
""" """
super().__init__() super().__init__(**kwargs)
self._user_audio_buffer = bytearray() self._user_audio_buffer = bytearray()
self._assistant_audio_buffer = bytearray() self._assistant_audio_buffer = bytearray()
self._num_channels = None self._num_channels = None
@@ -55,7 +53,7 @@ class AudioBufferProcessor(FrameProcessor):
with BytesIO() as buffer: with BytesIO() as buffer:
with wave.open(buffer, 'wb') as wf: with wave.open(buffer, 'wb') as wf:
wf.setnchannels(2) wf.setnchannels(2)
wf.setsampwidth(self._sample_rate // 8000) wf.setsampwidth(2)
wf.setframerate(self._sample_rate) wf.setframerate(self._sample_rate)
# Interleave the two audio streams # Interleave the two audio streams
max_length = max(len(self._user_audio_buffer), max_length = max(len(self._user_audio_buffer),
@@ -82,7 +80,7 @@ class AudioBufferProcessor(FrameProcessor):
async def process_frame(self, frame: Frame, direction: FrameDirection): async def process_frame(self, frame: Frame, direction: FrameDirection):
await super().process_frame(frame, direction) await super().process_frame(frame, direction)
if (isinstance(frame, AudioRawFrame) and self._sample_rate is None): if isinstance(frame, AudioRawFrame) and self._sample_rate is None:
self._sample_rate = frame.sample_rate self._sample_rate = frame.sample_rate
# include all audio from the user # include all audio from the user
@@ -95,7 +93,7 @@ class AudioBufferProcessor(FrameProcessor):
self._assistant_audio_buffer.extend(silence) self._assistant_audio_buffer.extend(silence)
# if the assistant is speaking, include all audio from the assistant, # if the assistant is speaking, include all audio from the assistant,
if (isinstance(frame, OutputAudioRawFrame)): if isinstance(frame, OutputAudioRawFrame):
self._assistant_audio_buffer.extend(frame.audio) self._assistant_audio_buffer.extend(frame.audio)
# do not push the user's audio frame, doing so will result in echo # do not push the user's audio frame, doing so will result in echo