something is up with transcription, trying to figure it out

This commit is contained in:
Moishe Lettvin
2024-01-10 17:12:40 -05:00
parent 7229fc806e
commit df1bbef653
3 changed files with 70 additions and 52 deletions

View File

@@ -5,10 +5,9 @@ from asyncio.queues import Queue
import re
from dailyai.output_queue import OutputQueueFrame, FrameType
from dailyai.services.azure_ai_services import AzureLLMService, AzureTTSService
from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
from dailyai.services.open_ai_services import OpenAILLMService, OpenAIImageGenService
from dailyai.services.azure_ai_services import AzureLLMService
from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
from dailyai.services.open_ai_services import OpenAIImageGenService
from dailyai.services.daily_transport_service import DailyTransportService
async def main(room_url):
@@ -89,6 +88,9 @@ async def main(room_url):
if participant["id"] == transport.my_participant_id:
return
# This will play the months in the order they're completed. The benefit
# is we'll have as little delay as possible before the first month, and
# likely no delay between months, but the months won't display in order.
for month_data_task in asyncio.as_completed(month_tasks):
data = await month_data_task
transport.output_queue.put(

View File

@@ -1,6 +1,8 @@
import argparse
import asyncio
import requests
import time
import urllib.parse
from dailyai.services.daily_transport_service import DailyTransportService
from dailyai.services.azure_ai_services import AzureLLMService, AzureTTSService
@@ -14,7 +16,7 @@ async def main(room_url:str, token):
transport = DailyTransportService(
room_url,
token,
"Say Two Things Bot",
"Respond bot",
1,
)
transport.mic_enabled = True
@@ -24,34 +26,37 @@ async def main(room_url:str, token):
llm = AzureLLMService()
tts = AzureTTSService()
transcribed_message = ""
transcription_timeout = None
"""
@transport.event_handler("on_participant_joined")
async def on_joined(transport, participant):
if participant["id"] == transport.my_participant_id:
return
# queue two pieces of speech: one specified as a text literal,
# and one generated by an llm. We'll kick off the llm first, and let
# it generate a response while we're speaking the literal string.
llm_response_task = asyncio.create_task(llm.run_llm(
[{"role": "system", "content": "tell the user a joke about llamas"}]
))
async for audio_chunk in tts.run_tts("My friend the LLM is now going to tell a joke about llamas."):
async for audio_chunk in tts.run_tts("If you say something, I will respond."):
transport.output_queue.put(OutputQueueFrame(FrameType.AUDIO_FRAME, audio_chunk))
llm_response = await llm_response_task
async for audio_chunk in tts.run_tts(llm_response):
transport.output_queue.put(OutputQueueFrame(FrameType.AUDIO_FRAME, audio_chunk))
# wait for the output queue to be empty, then leave the meeting
transport.output_queue.join()
transport.stop()
@transport.event_handler("on_transcription_message")
async def on_transcription_message(transport, message) -> None:
print("message received", message)
nonlocal transcribed_message
nonlocal transcription_timeout
print(message)
if message["session_id"] != transport.my_participant_id:
transcribed_message += message['text']
await transport.run()
print("message received", transcribed_message)
@transport.event_handler("on_transcription_error")
def on_transcription_error(transport, status) -> None:
print("transcription error", status)
@transport.event_handler("on_transcription_started")
def on_transcription_started(transport, status) -> None:
print("transcription started", status)
"""
#await transport.run()
transport.run()
if __name__ == "__main__":