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

minor cleanup

Browse Source
This commit is contained in:
Moishe Lettvin
2024-01-29 15:06:39 -05:00
parent d0bcddfd70
commit 4e9586595d
9 changed files with 40 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/dailyai/queue_aggregators.py
View File

@@ -34,6 +34,7 @@ class LLMContextAggregator(AIService):
bot_participant_id=None,
complete_sentences=True,
pass_through=True):
super().__init__()
self.messages = messages
self.bot_participant_id = bot_participant_id
self.role = role
@@ -82,5 +83,5 @@ class LLMAssistantContextAggregator(LLMContextAggregator):
self, messages: list[dict], bot_participant_id=None, complete_sentences=True
):
super().__init__(
messages, "assistan", bot_participant_id, complete_sentences, pass_through=True
messages, "assistant", bot_participant_id, complete_sentences, pass_through=True
)

11
src/dailyai/services/daily_transport_service.py
View File

@@ -391,10 +391,10 @@ class DailyTransportService(EventHandler):
all_audio_frames.extend(chunk)
b.extend(chunk)
l = len(b) - (len(b) % smallest_write_size)
if l:
self.mic.write_frames(bytes(b[:l]))
b = b[l:]
truncated_length: int = len(b) - (len(b) % smallest_write_size)
if truncated_length:
self.mic.write_frames(bytes(b[:truncated_length]))
b = b[truncated_length:]
elif isinstance(frame, ImageQueueFrame):
self._set_image(frame.image)
elif isinstance(frame, SpriteQueueFrame):
@@ -406,7 +406,8 @@ class DailyTransportService(EventHandler):
# if there are leftover audio bytes, write them now; failing to do so
# can cause static in the audio stream.
if len(b):
self.mic.write_frames(bytes(b))
truncated_length = len(b) - (len(b) % 160)
self.mic.write_frames(bytes(b[:truncated_length]))
b = bytearray()
if isinstance(frame, StartStreamQueueFrame):

1
src/samples/foundational/02-llm-say-one-thing.py
View File

@@ -1,6 +1,5 @@
import argparse
import asyncio
import logging
import aiohttp

8
src/samples/foundational/03-still-frame.py
View File

@@ -5,7 +5,7 @@ import aiohttp
from dailyai.queue_frame import TextQueueFrame
from dailyai.services.daily_transport_service import DailyTransportService
from dailyai.services.open_ai_services import OpenAIImageGenService
from dailyai.services.fal_ai_services import FalImageGenService
local_joined = False
participant_joined = False
@@ -25,14 +25,14 @@ async def main(room_url):
transport.camera_width = 1024
transport.camera_height = 1024
imagegen = OpenAIImageGenService(image_size="1024x1024", aiohttp_session=session)
imagegen = FalImageGenService(image_size="1024x1024", aiohttp_session=session)
image_task = asyncio.create_task(
imagegen.run_to_queue(
transport.send_queue, [
TextQueueFrame("a cat in the style of picasso")]))
@transport.event_handler("on_participant_joined")
async def on_participant_joined(transport, participant):
@transport.event_handler("on_first_other_participant_joined")
async def on_first_other_participant_joined(transport):
await image_task
await transport.run()

7
src/samples/foundational/04-utterance-and-speech.py
View File

@@ -40,11 +40,8 @@ async def main(room_url: str):
)
)
@transport.event_handler("on_participant_joined")
async def on_joined(transport, participant):
if participant["id"] == transport.my_participant_id:
return
@transport.event_handler("on_first_other_participant_joined")
async def on_first_other_participant_joined(transport):
await azure_tts.say("My friend the LLM is now going to tell a joke about llamas.", transport.send_queue)
async def buffer_to_send_queue():

14
src/samples/foundational/08-bots-arguing.py
View File

@@ -1,14 +1,10 @@
import aiohttp
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
from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
from dailyai.queue_frame import QueueFrame
async def main(room_url:str):
async with aiohttp.ClientSession() as session:
@@ -43,15 +39,17 @@ async def main(room_url:str):
# Run the LLMs synchronously for the back-and-forth
bot1_msg = await llm.run_llm(bot1_messages)
print(f"bot1_msg: {bot1_msg}")
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
if bot1_msg:
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
await tts1.say(bot1_msg, transport.send_queue)
bot2_msg = await llm.run_llm(bot2_messages)
print(f"bot2_msg: {bot2_msg}")
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
if bot2_msg:
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
await tts2.say(bot2_msg, transport.send_queue)

17
src/samples/foundational/08a-characters-arguing.py
View File

@@ -1,16 +1,12 @@
import aiohttp
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
from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
from dailyai.services.fal_ai_services import FalImageGenService
from dailyai.services.open_ai_services import OpenAIImageGenService
from dailyai.queue_frame import QueueFrame, AudioQueueFrame, ImageQueueFrame
from dailyai.queue_frame import AudioQueueFrame, ImageQueueFrame
async def main(room_url:str):
async with aiohttp.ClientSession() as session:
@@ -34,7 +30,6 @@ async def main(room_url:str):
tts1 = AzureTTSService()
tts2 = ElevenLabsTTSService(session)
dalle = FalImageGenService(image_size="1024x1024", aiohttp_session=session)
# dalle = OpenAIImageGenService(image_size="1024x1024")
bot1_messages = [
{"role": "system", "content": "You are a stern librarian. You strongly believe that a hot dog is a sandwich. Start by stating this fact in a few sentences, then be prepared to debate this with the user. You shouldn't ever compromise on the fundamental truth that a hot dog is a sandwich. Your responses should only be a few sentences long."},
@@ -47,8 +42,9 @@ async def main(room_url:str):
# Run the LLMs synchronously for the back-and-forth
bot1_msg = await llm.run_llm(bot1_messages)
print(f"bot1_msg: {bot1_msg}")
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
if bot1_msg:
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
all_audio = bytearray()
async for audio in tts1.run_tts(bot1_msg):
@@ -60,8 +56,9 @@ async def main(room_url:str):
# Run the LLMs synchronously for the back-and-forth
bot2_msg = await llm.run_llm(bot2_messages)
print(f"bot2_msg: {bot2_msg}")
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
if bot2_msg:
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
all_audio = bytearray()
async for audio in tts2.run_tts(bot2_msg):

16
src/samples/foundational/08b-debate-generator.py
View File

@@ -1,16 +1,12 @@
import aiohttp
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
from dailyai.services.elevenlabs_ai_service import ElevenLabsTTSService
from dailyai.services.fal_ai_services import FalImageGenService
from dailyai.services.open_ai_services import OpenAIImageGenService
from dailyai.queue_frame import QueueFrame, AudioQueueFrame, ImageQueueFrame
from dailyai.queue_frame import AudioQueueFrame, ImageQueueFrame
async def main(room_url:str):
async with aiohttp.ClientSession() as session:
@@ -57,8 +53,9 @@ async def main(room_url:str):
# Run the LLMs synchronously for the back-and-forth
bot1_msg = await llm.run_llm(bot1_messages)
print(f"bot1_msg: {bot1_msg}")
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
if bot1_msg:
bot1_messages.append({"role": "assistant", "content": bot1_msg})
bot2_messages.append({"role": "user", "content": bot1_msg})
all_audio = bytearray()
async for audio in tts1.run_tts(bot1_msg):
@@ -70,8 +67,9 @@ async def main(room_url:str):
# Run the LLMs synchronously for the back-and-forth
bot2_msg = await llm.run_llm(bot2_messages)
print(f"bot2_msg: {bot2_msg}")
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
if bot2_msg:
bot2_messages.append({"role": "assistant", "content": bot2_msg})
bot1_messages.append({"role": "user", "content": bot2_msg})
all_audio = bytearray()
async for audio in tts2.run_tts(bot2_msg):

2
src/samples/foundational/10-wake-word.py
View File

@@ -73,7 +73,7 @@ class TranscriptFilter(AIService):
class NameCheckFilter(AIService):
def __init__(self, names=None):
def __init__(self, names:list[str]):
self.names = names
self.sentence = ""