Added async OpenAI services (#1)

* added async openai services

* added async openai services

* added Deepgram service with 05 example

* modernized the 'say one thing' example

* async all the things

* cleanup and user greeting

* more cleanup

src/samples/simple-sample/simple-sample.py

@@ -13,6 +13,7 @@ from dailyai.orchestrator import OrchestratorConfig, Orchestrator
 from dailyai.message_handler.message_handler import MessageHandler
 from dailyai.services.ai_services import AIServiceConfig
 from dailyai.services.azure_ai_services import AzureImageGenService, AzureTTSService, AzureLLMService
+from dailyai.services.deepgram_ai_services import DeepgramTTSService
 
 def add_bot_to_room(room_url, token, expiration) -> None:
 

src/samples/theoretical-to-real/01a-greet-user.py

@@ -0,0 +1,55 @@
+import asyncio
+import time
+from typing import AsyncGenerator
+
+from dailyai.output_queue import OutputQueueFrame, FrameType
+from dailyai.services.daily_transport_service import DailyTransportService
+from dailyai.services.azure_ai_services import AzureTTSService
+from dailyai.services.deepgram_ai_services import DeepgramTTSService
+
+async def main(room_url):
+    # create a transport service object using environment variables for
+    # the transport service's API key, room url, and any other configuration.
+    # services can all define and document the environment variables they use.
+    # services all also take an optional config object that is used instead of
+    # environment variables.
+    #
+    # the abstract transport service APIs presumably can map pretty closely
+    # to the daily-python basic API
+    meeting_duration_minutes = 1
+    transport = DailyTransportService(
+        room_url,
+        None,
+        "Greeter",
+        meeting_duration_minutes,
+    )
+    transport.mic_enabled = True
+
+    # similarly, create a tts service
+    tts = DeepgramTTSService()
+
+    # Get the generator for the audio. This will start running in the background,
+    # and when we ask the generator for its items, we'll get what it's generated.
+
+    # Register an event handler so we can play the audio when the participant joins.
+    print("settting up handler")
+    @transport.event_handler("on_participant_joined")
+    async def on_participant_joined(transport, participant):
+        print(f"participant joined: {participant['info']['userName']}")
+        if participant["info"]["isLocal"]:
+            return
+        audio_generator: AsyncGenerator[bytes, None] = tts.run_tts(f"Hello there, {participant['info']['userName']}!")
+
+        async for audio in audio_generator:
+            transport.output_queue.put(OutputQueueFrame(FrameType.AUDIO_FRAME, audio))
+
+    print("setting up call state handler")
+    @transport.event_handler("on_call_state_updated")
+    async def on_call_joined(transport, state):
+        print(f"call state callback: {state}")
+    
+    await transport.run()
+
+
+if __name__ == "__main__":
+    asyncio.run(main("https://chad-hq.daily.co/howdy"))

src/samples/theoretical-to-real/05-sync-speech-and-text.py

@@ -1,7 +1,9 @@
 import asyncio
 
 from dailyai.output_queue import OutputQueueFrame, FrameType
-from dailyai.services.azure_ai_services import AzureLLMService, AzureTTSService, AzureImageGenServiceREST
+from dailyai.services.azure_ai_services import AzureTTSService
+from dailyai.services.open_ai_services import OpenAILLMService, OpenAIImageGenService
+from dailyai.services.deepgram_ai_services import DeepgramTTSService
 from dailyai.services.daily_transport_service import DailyTransportService
 
 async def main(room_url, token):
@@ -23,9 +25,9 @@ async def main(room_url, token):
     transport.camera_width = 1024
     transport.camera_height = 1024
 
-    llm = AzureLLMService()
-    tts = AzureTTSService()
-    dalle = AzureImageGenServiceREST()
+    llm = OpenAILLMService()
+    tts = DeepgramTTSService()
+    dalle = OpenAIImageGenService()
 
     async def get_all_audio(text):
         all_audio = bytearray()
@@ -44,7 +46,7 @@ async def main(room_url, token):
                 }
             ]
         )
-        print(f"got llm for {month}")
+        print(f"got llm for {month}, {inference_text}")
 
         (image, audio) = await asyncio.gather(
             *[dalle.run_image_gen(inference_text, "1024x1024"), get_all_audio(inference_text)]
@@ -90,4 +92,4 @@ async def main(room_url, token):
     print("Done")
 
 if __name__=="__main__":
-    asyncio.run(main("https://moishe.daily.co/Lettvins", None))
+    asyncio.run(main("https://chad-hq.daily.co/howdy", None))