services(riva): cleanup
This commit is contained in:
vipyne
@@ -1,94 +0,0 @@
|
||||
#
|
||||
# Copyright (c) 2024, Daily
|
||||
#
|
||||
# SPDX-License-Identifier: BSD 2-Clause License
|
||||
#
|
||||
|
||||
import asyncio
|
||||
import os
|
||||
import sys
|
||||
|
||||
import aiohttp
|
||||
from dotenv import load_dotenv
|
||||
from loguru import logger
|
||||
from runner import configure
|
||||
|
||||
from pipecat.audio.vad.silero import SileroVADAnalyzer
|
||||
from pipecat.frames.frames import LLMMessagesFrame
|
||||
from pipecat.pipeline.pipeline import Pipeline
|
||||
from pipecat.pipeline.runner import PipelineRunner
|
||||
from pipecat.pipeline.task import PipelineParams, PipelineTask
|
||||
from pipecat.processors.aggregators.openai_llm_context import OpenAILLMContext
|
||||
|
||||
from pipecat.services.deepgram import DeepgramSTTService
|
||||
from pipecat.services.riva import FastpitchTTSService, ParakeetSTTService
|
||||
|
||||
from pipecat.services.openai import OpenAILLMService
|
||||
from pipecat.transports.services.daily import DailyParams, DailyTransport
|
||||
|
||||
load_dotenv(override=True)
|
||||
|
||||
logger.remove(0)
|
||||
logger.add(sys.stderr, level="DEBUG")
|
||||
|
||||
|
||||
async def main():
|
||||
async with aiohttp.ClientSession() as session:
|
||||
(room_url, _) = await configure(session)
|
||||
|
||||
transport = DailyTransport(
|
||||
room_url,
|
||||
None,
|
||||
"Respond bot",
|
||||
DailyParams(
|
||||
audio_out_enabled=True,
|
||||
vad_enabled=True,
|
||||
vad_analyzer=SileroVADAnalyzer(),
|
||||
vad_audio_passthrough=True,
|
||||
),
|
||||
)
|
||||
|
||||
# stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
|
||||
stt = ParakeetSTTService(api_key=os.getenv("NVIDIA_API_KEY"))
|
||||
|
||||
tts = FastpitchTTSService(api_key=os.getenv("NVIDIA_API_KEY"))
|
||||
|
||||
llm = OpenAILLMService(api_key=os.getenv("OPENAI_API_KEY"), model="gpt-4o")
|
||||
|
||||
messages = [
|
||||
{
|
||||
"role": "system",
|
||||
"content": "You are a helpful LLM in a WebRTC call. Your goal is to demonstrate your capabilities in a succinct way. Your output will be converted to audio so don't include special characters in your answers. Respond to what the user said in a creative and helpful way.",
|
||||
},
|
||||
]
|
||||
|
||||
context = OpenAILLMContext(messages)
|
||||
context_aggregator = llm.create_context_aggregator(context)
|
||||
|
||||
pipeline = Pipeline(
|
||||
[
|
||||
transport.input(), # Transport user input
|
||||
stt, # STT
|
||||
context_aggregator.user(), # User responses
|
||||
llm, # LLM
|
||||
tts, # TTS
|
||||
transport.output(), # Transport bot output
|
||||
context_aggregator.assistant(), # Assistant spoken responses
|
||||
]
|
||||
)
|
||||
|
||||
task = PipelineTask(pipeline, PipelineParams(allow_interruptions=True))
|
||||
|
||||
@transport.event_handler("on_first_participant_joined")
|
||||
async def on_first_participant_joined(transport, participant):
|
||||
# Kick off the conversation.
|
||||
messages.append({"role": "system", "content": "Please introduce yourself to the user."})
|
||||
await task.queue_frames([LLMMessagesFrame(messages)])
|
||||
|
||||
runner = PipelineRunner()
|
||||
|
||||
await runner.run(task)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
asyncio.run(main())
|
||||
@@ -5,11 +5,7 @@
|
||||
#
|
||||
|
||||
import asyncio
|
||||
<<<<<<< HEAD
|
||||
from typing import AsyncGenerator, Optional
|
||||
=======
|
||||
from typing import AsyncGenerator, List, Optional, Union, Iterator
|
||||
>>>>>>> 07e3942e (add nvidia riva - fastpitch)
|
||||
|
||||
from loguru import logger
|
||||
from pydantic.main import BaseModel
|
||||
@@ -17,10 +13,6 @@ from pydantic.main import BaseModel
|
||||
from pipecat.frames.frames import (
|
||||
CancelFrame,
|
||||
EndFrame,
|
||||
<<<<<<< HEAD
|
||||
=======
|
||||
ErrorFrame,
|
||||
>>>>>>> 07e3942e (add nvidia riva - fastpitch)
|
||||
Frame,
|
||||
InterimTranscriptionFrame,
|
||||
StartFrame,
|
||||
@@ -46,11 +38,8 @@ except ModuleNotFoundError as e:
|
||||
|
||||
class FastpitchTTSService(TTSService):
|
||||
class InputParams(BaseModel):
|
||||
<<<<<<< HEAD
|
||||
language: Optional[Language] = Language.EN_US
|
||||
=======
|
||||
language: Optional[str] = "en-US"
|
||||
>>>>>>> 07e3942e (add nvidia riva - fastpitch)
|
||||
quality: Optional[int] = 20
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
@@ -58,19 +47,18 @@ class FastpitchTTSService(TTSService):
|
||||
api_key: str,
|
||||
server: str = "grpc.nvcf.nvidia.com:443",
|
||||
voice_id: str = "English-US.Female-1",
|
||||
sample_rate_hz: int = 24000,
|
||||
sample_rate: int = 24000,
|
||||
# nvidia riva calls this 'function-id'
|
||||
model: str = "0149dedb-2be8-4195-b9a0-e57e0e14f972",
|
||||
params: InputParams = InputParams(),
|
||||
quality: int = 20,
|
||||
**kwargs,
|
||||
):
|
||||
super().__init__(sample_rate=sample_rate_hz, **kwargs)
|
||||
super().__init__(sample_rate=sample_rate, **kwargs)
|
||||
self._api_key = api_key
|
||||
self._voice_id = voice_id
|
||||
self._sample_rate_hz = sample_rate_hz
|
||||
self._sample_rate = sample_rate
|
||||
self._language_code = params.language
|
||||
self.quality = quality
|
||||
self.quality = params.quality
|
||||
|
||||
self.set_model_name("fastpitch-hifigan-tts")
|
||||
self.set_voice(voice_id)
|
||||
@@ -81,16 +69,16 @@ class FastpitchTTSService(TTSService):
|
||||
]
|
||||
auth = riva.client.Auth(None, True, server, metadata)
|
||||
|
||||
self.service = riva.client.SpeechSynthesisService(auth)
|
||||
self._service = riva.client.SpeechSynthesisService(auth)
|
||||
|
||||
async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
|
||||
def read_audio_responses():
|
||||
try:
|
||||
responses = self.service.synthesize_online(
|
||||
responses = self._service.synthesize_online(
|
||||
text,
|
||||
self._voice_id,
|
||||
self._language_code,
|
||||
sample_rate_hz=self._sample_rate_hz,
|
||||
sample_rate_hz=self._sample_rate,
|
||||
audio_prompt_file=None,
|
||||
quality=self.quality,
|
||||
custom_dictionary={},
|
||||
@@ -112,7 +100,7 @@ class FastpitchTTSService(TTSService):
|
||||
|
||||
frame = TTSAudioRawFrame(
|
||||
audio=resp.audio,
|
||||
sample_rate=self._sample_rate_hz,
|
||||
sample_rate=self._sample_rate,
|
||||
num_channels=1,
|
||||
)
|
||||
yield frame
|
||||
@@ -150,7 +138,7 @@ class ParakeetSTTService(STTService):
|
||||
self._stop_history_eou = -1
|
||||
self._stop_threshold_eou = -1.0
|
||||
self._custom_configuration = ""
|
||||
self._sample_rate_hz: int = 16000
|
||||
self._sample_rate: int = 16000
|
||||
|
||||
self.set_model_name("parakeet-ctc-1.1b-asr")
|
||||
|
||||
@@ -171,7 +159,7 @@ class ParakeetSTTService(STTService):
|
||||
profanity_filter=self._profanity_filter,
|
||||
enable_automatic_punctuation=self._automatic_punctuation,
|
||||
verbatim_transcripts=not self._no_verbatim_transcripts,
|
||||
sample_rate_hertz=self._sample_rate_hz,
|
||||
sample_rate_hertz=self._sample_rate,
|
||||
audio_channel_count=1,
|
||||
),
|
||||
interim_results=True,
|
||||
|
||||
Reference in New Issue
Block a user