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

fix: Use audio_in_sample_rate to deserialize data for TwilioFrameSerializer

Browse Source
This commit is contained in:
Mark Backman
2025-02-06 09:55:15 -05:00
parent 99d3227ff5
commit ba59736f87
2 changed files with 16 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
CHANGELOG.md
View File

@@ -5,6 +5,14 @@ All notable changes to **Pipecat** will be documented in this file.
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/), The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
## [Unreleased]
### Fixed
- Fixed an issue with the `TwilioFrameSerializer` where
`frame.audio_out_sample_rate` was incorrectly used in place of
`frame.audio_in_sample_rate`.
## [0.0.55] - 2025-02-05 ## [0.0.55] - 2025-02-05
### Added ### Added

13
src/pipecat/serializers/twilio.py
View File

@@ -27,15 +27,15 @@ from pipecat.serializers.base_serializer import FrameSerializer, FrameSerializer
class TwilioFrameSerializer(FrameSerializer): class TwilioFrameSerializer(FrameSerializer):
class InputParams(BaseModel): class InputParams(BaseModel):
twilio_sample_rate: Optional[int] = None twilio_sample_rate: Optional[int] = None # Default Twilio rate (8kHz)
sample_rate: Optional[int] = None sample_rate: Optional[int] = None # Pipeline input rate
def __init__(self, stream_sid: str, params: InputParams = InputParams()): def __init__(self, stream_sid: str, params: InputParams = InputParams()):
self._stream_sid = stream_sid self._stream_sid = stream_sid
self._params = params self._params = params
self._twilio_sample_rate = 0 self._twilio_sample_rate = 0 # Fixed rate for Twilio (8kHz)
self._sample_rate = 0 self._sample_rate = 0 # Pipeline input rate
self._resampler = create_default_resampler() self._resampler = create_default_resampler()
@@ -44,8 +44,9 @@ class TwilioFrameSerializer(FrameSerializer):
return FrameSerializerType.TEXT return FrameSerializerType.TEXT
async def setup(self, frame: StartFrame): async def setup(self, frame: StartFrame):
# Configure rates for input path: Twilio (8kHz μ-law) -> Pipeline (PCM)
self._twilio_sample_rate = self._params.twilio_sample_rate or frame.audio_in_sample_rate self._twilio_sample_rate = self._params.twilio_sample_rate or frame.audio_in_sample_rate
self._sample_rate = self._params.sample_rate or frame.audio_out_sample_rate self._sample_rate = self._params.sample_rate or frame.audio_in_sample_rate
async def serialize(self, frame: Frame) -> str | bytes | None: async def serialize(self, frame: Frame) -> str | bytes | None:
if isinstance(frame, StartInterruptionFrame): if isinstance(frame, StartInterruptionFrame):
@@ -54,6 +55,7 @@ class TwilioFrameSerializer(FrameSerializer):
elif isinstance(frame, AudioRawFrame): elif isinstance(frame, AudioRawFrame):
data = frame.audio data = frame.audio
# Output: Convert PCM at frame's rate to 8kHz μ-law for Twilio
serialized_data = await pcm_to_ulaw( serialized_data = await pcm_to_ulaw(
data, frame.sample_rate, self._twilio_sample_rate, self._resampler data, frame.sample_rate, self._twilio_sample_rate, self._resampler
) )
@@ -75,6 +77,7 @@ class TwilioFrameSerializer(FrameSerializer):
payload_base64 = message["media"]["payload"] payload_base64 = message["media"]["payload"]
payload = base64.b64decode(payload_base64) payload = base64.b64decode(payload_base64)
# Input: Convert Twilio's 8kHz μ-law to PCM at pipeline input rate
deserialized_data = await ulaw_to_pcm( deserialized_data = await ulaw_to_pcm(
payload, self._twilio_sample_rate, self._sample_rate, self._resampler payload, self._twilio_sample_rate, self._sample_rate, self._resampler
) )