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

WIP: environment cleanup (#19)

Browse Source 
* removed env var usage from SDK services

* started consolidating configure.py

* 1–3 work

* cleaned up the rest

* more cleanup

* cleanup and 05 tinkering

* made fal keys optional
This commit is contained in:
chadbailey59
2024-02-06 15:07:16 -06:00
committed by GitHub
parent 9d5ad5675c
commit 70d07b6ea2
24 changed files with 475 additions and 804 deletions
Download Patch File Download Diff File Expand all files Collapse all files

56
src/dailyai/services/azure_ai_services.py
View File

@@ -17,13 +17,10 @@ from azure.cognitiveservices.speech import SpeechSynthesizer, SpeechConfig, Resu
class AzureTTSService(TTSService):
def __init__(self, speech_key=None, speech_region=None):
def __init__(self, *, api_key, region):
super().__init__()
speech_key = speech_key or os.getenv("AZURE_SPEECH_SERVICE_KEY")
speech_region = speech_region or os.getenv("AZURE_SPEECH_SERVICE_REGION")
self.speech_config = SpeechConfig(subscription=speech_key, region=speech_region)
self.speech_config = SpeechConfig(subscription=api_key, region=region)
self.speech_synthesizer = SpeechSynthesizer(
speech_config=self.speech_config, audio_config=None)
@@ -51,25 +48,13 @@ class AzureTTSService(TTSService):
class AzureLLMService(LLMService):
def __init__(self, api_key=None, azure_endpoint=None, api_version=None, model=None):
def __init__(self, *, api_key, endpoint, api_version="2023-12-01-preview", model):
super().__init__()
api_key = api_key or os.getenv("AZURE_CHATGPT_KEY")
self._model: str = model
azure_endpoint = azure_endpoint or os.getenv("AZURE_CHATGPT_ENDPOINT")
if not azure_endpoint:
raise Exception(
"No azure endpoint specified for Azure LLM, please set AZURE_CHATGPT_ENDPOINT in the environment or pass it to the AzureLLMService constructor")
model: str | None = model or os.getenv("AZURE_CHATGPT_DEPLOYMENT_ID")
if not model:
raise Exception(
"No model specified for Azure LLM, please set AZURE_CHATGPT_DEPLOYMENT_ID in the environment or pass it to the AzureLLMService constructor")
self.model: str = model
api_version = api_version or "2023-12-01-preview"
self.client = AsyncAzureOpenAI(
self._client = AsyncAzureOpenAI(
api_key=api_key,
azure_endpoint=azure_endpoint,
azure_endpoint=endpoint,
api_version=api_version,
)
@@ -77,7 +62,7 @@ class AzureLLMService(LLMService):
messages_for_log = json.dumps(messages)
self.logger.debug(f"Generating chat via azure: {messages_for_log}")
chunks = await self.client.chat.completions.create(model=self.model, stream=True, messages=messages)
chunks = await self._client.chat.completions.create(model=self._model, stream=True, messages=messages)
async for chunk in chunks:
if len(chunk.choices) == 0:
continue
@@ -89,7 +74,7 @@ class AzureLLMService(LLMService):
messages_for_log = json.dumps(messages)
self.logger.debug(f"Generating chat via azure: {messages_for_log}")
response = await self.client.chat.completions.create(model=self.model, stream=False, messages=messages)
response = await self._client.chat.completions.create(model=self._model, stream=False, messages=messages)
if response and len(response.choices) > 0:
return response.choices[0].message.content
else:
@@ -100,17 +85,19 @@ class AzureImageGenServiceREST(ImageGenService):
def __init__(
self,
*,
api_version="2023-06-01-preview",
image_size: str,
aiohttp_session: aiohttp.ClientSession,
api_key=None,
azure_endpoint=None,
api_version=None,
model=None):
api_key,
endpoint,
model):
super().__init__(image_size=image_size)
self._api_key = api_key or os.getenv("AZURE_DALLE_KEY")
self._azure_endpoint = azure_endpoint or os.getenv("AZURE_DALLE_ENDPOINT")
self._api_version = api_version or "2023-06-01-preview"
self._model = model or os.getenv("AZURE_DALLE_DEPLOYMENT_ID")
self._api_key = api_key
self._azure_endpoint = endpoint
self._api_version = api_version
self._model = model
self._aiohttp_session = aiohttp_session
async def run_image_gen(self, sentence) -> tuple[str, bytes]:
@@ -125,8 +112,11 @@ class AzureImageGenServiceREST(ImageGenService):
async with self._aiohttp_session.post(
url, headers=headers, json=body
) as submission:
print(f"submission: {submission}")
# We never get past this line, because this header isn't
# defined on a 429 response, but something is eating our exceptions!
operation_location = submission.headers['operation-location']
print(f"submission status: {submission.status}")
status = ""
attempts_left = 120
json_response = None
@@ -145,9 +135,9 @@ class AzureImageGenServiceREST(ImageGenService):
image_url = json_response["result"]["data"][0]["url"] if json_response else None
if not image_url:
raise Exception("Image generation failed")
# Load the image from the url
async with self._aiohttp_session.get(image_url) as response:
image_stream = io.BytesIO(await response.content.read())
image = Image.open(image_stream)
print("i got an image file!")
return (image_url, image.tobytes())

8
src/dailyai/services/deepgram_ai_services.py
View File

@@ -9,11 +9,11 @@ from dailyai.services.ai_services import TTSService
class DeepgramTTSService(TTSService):
def __init__(self, aiohttp_session, speech_key=None, voice=None):
def __init__(self, *, aiohttp_session, api_key, voice="alpha-asteria-en-v2"):
super().__init__()
self._voice = voice or os.getenv("DEEPGRAM_VOICE") or "alpha-asteria-en-v2"
self._speech_key = speech_key or os.getenv("DEEPGRAM_API_KEY")
self._voice = voice
self._api_key = api_key
self._aiohttp_session = aiohttp_session
def get_mic_sample_rate(self):
@@ -23,7 +23,7 @@ class DeepgramTTSService(TTSService):
self.logger.info(f"Running deepgram tts for {sentence}")
base_url = "https://api.beta.deepgram.com/v1/speak"
request_url = f"{base_url}?model={self._voice}&encoding=linear16&container=none&sample_rate=16000"
headers = {"authorization": f"token {self._speech_key}"}
headers = {"authorization": f"token {self._api_key}"}
body = {"text": sentence}
async with self._aiohttp_session.post(request_url, headers=headers, json=body) as r:
async for data in r.content:

9
src/dailyai/services/elevenlabs_ai_service.py
View File

@@ -12,14 +12,15 @@ class ElevenLabsTTSService(TTSService):
def __init__(
self,
*,
aiohttp_session: aiohttp.ClientSession,
api_key=None,
voice_id=None,
api_key,
voice_id,
):
super().__init__()
self._api_key = api_key or os.getenv("ELEVENLABS_API_KEY")
self._voice_id = voice_id or os.getenv("ELEVENLABS_VOICE_ID")
self._api_key = api_key
self._voice_id = voice_id
self._aiohttp_session = aiohttp_session
async def run_tts(self, sentence) -> AsyncGenerator[bytes, None]:

8
src/dailyai/services/fal_ai_services.py
View File

@@ -2,18 +2,22 @@ import fal
import aiohttp
import asyncio
import io
import os
import json
from PIL import Image
from dailyai.services.ai_services import LLMService, TTSService, ImageGenService
# Fal expects FAL_KEY_ID and FAL_KEY_SECRET to be set in the env
class FalImageGenService(ImageGenService):
def __init__(self, image_size, aiohttp_session: aiohttp.ClientSession):
def __init__(self, *, image_size, aiohttp_session: aiohttp.ClientSession, key_id=None, key_secret=None):
super().__init__(image_size)
self._aiohttp_session = aiohttp_session
if key_id:
os.environ["FAL_KEY_ID"] = key_id
if key_secret:
os.environ["FAL_KEY_SECRET"] = key_secret
async def run_image_gen(self, sentence) -> tuple[str, bytes]:
def get_image_url(sentence, size):

27
src/dailyai/services/open_ai_services.py
View File

@@ -13,26 +13,24 @@ from dailyai.services.ai_services import AIService, TTSService, LLMService, Imag
class OpenAILLMService(LLMService):
def __init__(self, api_key=None, model=None):
def __init__(self, *, api_key, model="gpt-4"):
super().__init__()
api_key = api_key or os.getenv("OPEN_AI_KEY")
self.model = model or os.getenv("OPEN_AI_LLM_MODEL") or "gpt-4"
self.client = AsyncOpenAI(api_key=api_key)
self._model = model
self._client = AsyncOpenAI(api_key=api_key)
async def get_response(self, messages, stream):
return await self.client.chat.completions.create(
return await self._client.chat.completions.create(
stream=stream,
messages=messages,
model=self.model
model=self._model
)
async def run_llm_async(self, messages) -> AsyncGenerator[str, None]:
messages_for_log = json.dumps(messages)
self.logger.debug(f"Generating chat via openai: {messages_for_log}")
response = await self.get_response(messages, stream=True)
for chunk in response:
chunks = await self._client.chat.completions.create(model=self._model, stream=True, messages=messages)
async for chunk in chunks:
if len(chunk.choices) == 0:
continue
@@ -43,7 +41,7 @@ class OpenAILLMService(LLMService):
messages_for_log = json.dumps(messages)
self.logger.debug(f"Generating chat via openai: {messages_for_log}")
response = await self.get_response(messages, stream=False)
response = await self._client.chat.completions.create(model=self._model, stream=False, messages=messages)
if response and len(response.choices) > 0:
return response.choices[0].message.content
else:
@@ -54,14 +52,15 @@ class OpenAIImageGenService(ImageGenService):
def __init__(
self,
*,
image_size: str,
aiohttp_session: aiohttp.ClientSession,
api_key=None,
model=None,
api_key,
model="dall-e-3",
):
super().__init__(image_size=image_size)
api_key = api_key or os.getenv("OPEN_AI_KEY")
self._model = model or os.getenv("OPEN_AI_IMAGE_MODEL") or "dall-e-3"
self._model = model
print(f"api key: {api_key}")
self._client = AsyncOpenAI(api_key=api_key)
self._aiohttp_session = aiohttp_session