Add sample 04-

src/samples/theoretical-to-real/04-utterance-and-speech.py

@@ -1,33 +1,52 @@
 import argparse
 import asyncio
+import re
 
-from dailyai.output_queue import OutputQueueFrame, FrameType
 from dailyai.services.daily_transport_service import DailyTransportService
-from dailyai.services.open_ai_services import OpenAIImageGenService
+from dailyai.services.azure_ai_services import AzureLLMService, AzureTTSService
+from dailyai.output_queue import OutputQueueFrame, FrameType
 
-local_joined = False
-participant_joined = False
+async def main(room_url:str):
+    global transport
+    global llm
+    global tts
 
-async def main(room_url):
-    meeting_duration_minutes = 1
     transport = DailyTransportService(
         room_url,
         None,
-        "Show a still frame image",
-        meeting_duration_minutes,
+        "Say Two Things Bot",
+        1,
     )
-    transport.mic_enabled = False
-    transport.camera_enabled = True
-    transport.camera_width = 1024
-    transport.camera_height = 1024
+    transport.mic_enabled = True
+    transport.mic_sample_rate = 16000
+    transport.camera_enabled = False
 
-    imagegen = OpenAIImageGenService()
-    image_task = asyncio.create_task(imagegen.run_image_gen("a cat in the style of picasso", "1024x1024"))
+    llm = AzureLLMService()
+    tts = AzureTTSService()
 
     @transport.event_handler("on_participant_joined")
-    async def on_participant_joined(transport, participant):
-        (_, image_bytes) = await image_task
-        transport.output_queue.put(OutputQueueFrame(FrameType.IMAGE_FRAME, image_bytes))
+    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."):
+            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()
 
     await transport.run()