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

fix class heirarchy and audio chunking

Browse Source
This commit is contained in:
Kwindla Hultman Kramer
2025-03-26 07:54:57 -07:00
committed by Mark Backman
parent 060bb4c26b
commit 406f5a395b
2 changed files with 8 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
examples/foundational/07y-interruptible-groq.py
View File

@@ -48,7 +48,7 @@ async def main():
llm = GroqLLMService(api_key=os.getenv("GROQ_API_KEY"), model="llama-3.3-70b-versatile") llm = GroqLLMService(api_key=os.getenv("GROQ_API_KEY"), model="llama-3.3-70b-versatile")
tts = GroqTTSService(api_key=os.getenv("GROQ_API_KEY"), voice_id="Atlas-PlayAI") tts = GroqTTSService(api_key=os.getenv("GROQ_API_KEY"))
messages = [ messages = [
{ {

27
src/pipecat/services/groq.py
View File

@@ -12,7 +12,7 @@ from loguru import logger
from pydantic import BaseModel from pydantic import BaseModel
from pipecat.frames.frames import Frame, TTSAudioRawFrame, TTSStartedFrame, TTSStoppedFrame from pipecat.frames.frames import Frame, TTSAudioRawFrame, TTSStartedFrame, TTSStoppedFrame
from pipecat.services.ai_services import InterruptibleTTSService from pipecat.services.ai_services import TTSService
from pipecat.services.base_whisper import BaseWhisperSTTService, Transcription from pipecat.services.base_whisper import BaseWhisperSTTService, Transcription
from pipecat.services.openai import OpenAILLMService from pipecat.services.openai import OpenAILLMService
from pipecat.transcriptions.language import Language from pipecat.transcriptions.language import Language
@@ -104,7 +104,7 @@ class GroqSTTService(BaseWhisperSTTService):
return await self._client.audio.transcriptions.create(**kwargs) return await self._client.audio.transcriptions.create(**kwargs)
class GroqTTSService(InterruptibleTTSService): class GroqTTSService(TTSService):
class InputParams(BaseModel): class InputParams(BaseModel):
language: Optional[Language] = Language.EN language: Optional[Language] = Language.EN
speed: Optional[float] = 1.0 speed: Optional[float] = 1.0
@@ -117,7 +117,7 @@ class GroqTTSService(InterruptibleTTSService):
output_format: str = "wav", output_format: str = "wav",
params: InputParams = InputParams(), params: InputParams = InputParams(),
model_name: str = "playai-tts", model_name: str = "playai-tts",
voice_id: str = "Atlas-PlayAI", voice_id: str = "Celeste-PlayAI",
**kwargs, **kwargs,
): ):
super().__init__( super().__init__(
@@ -149,28 +149,15 @@ class GroqTTSService(InterruptibleTTSService):
input=text, input=text,
) )
async for data in response.iter_bytes(4096): async for data in response.iter_bytes():
if measuring_ttfb: if measuring_ttfb:
await self.stop_ttfb_metrics() await self.stop_ttfb_metrics()
measuring_ttfb = False measuring_ttfb = False
# remove wav header if present # remove wav header if present
if data.startswith(b"RIFF"): if data.startswith(b"RIFF"):
continue data = data[44:]
if len(data) == 0:
continue
yield TTSAudioRawFrame(data, 48000, 1) yield TTSAudioRawFrame(data, 48000, 1)
yield TTSStoppedFrame() yield TTSStoppedFrame()
async def _connect(self) -> None:
pass
async def _disconnect(self) -> None:
pass
async def _connect_websocket(self) -> None:
pass
async def _disconnect_websocket(self) -> None:
pass
async def _receive_messages(self) -> None:
pass