updated examples to follow APIs in the latest version (#1393)

Co-authored-by: Jayesh Parmar <[email protected]>

Author: davidzhao

examples/multimodal-agent/openai_agent.py

@@ -68,24 +68,20 @@ async def get_weather(
     #     fnc_ctx=fnc_ctx,
     # )
 
-    # create a chat context with chat history
+    # create a chat context with chat history, these will be synchronized with the server
+    # upon session establishment
     chat_ctx = llm.ChatContext()
-    chat_ctx.append(text="I'm planning a trip to Paris next month.", role="user")
-    chat_ctx.append(
-        text="How exciting! Paris is a beautiful city. I'd be happy to suggest some must-visit places and help you plan your trip.",
-        role="assistant",
-    )
-    chat_ctx.append(text="What are the must-visit places in Paris?", role="user")
-    chat_ctx.append(
-        text="The must-visit places in Paris are the Eiffel Tower, Louvre Museum, Notre-Dame Cathedral, and Montmartre.",
-        role="assistant",
-    )
+    # chat_ctx.append(text="I'm planning a trip to Paris next month.", role="user")
+    # chat_ctx.append(
+    #     text="How exciting! Paris is a beautiful city. I'd be happy to suggest some must-visit places and help you plan your trip.",
+    #     role="assistant",
+    # )
 
     agent = multimodal.MultimodalAgent(
         model=openai.realtime.RealtimeModel(
             voice="alloy",
             temperature=0.8,
-            instructions="You are a helpful assistant",
+            instructions="You are a helpful assistant, greet the user and help them with their trip planning",
             turn_detection=openai.realtime.ServerVadOptions(
                 threshold=0.6, prefix_padding_ms=200, silence_duration_ms=500
             ),
@@ -94,6 +90,7 @@ async def get_weather(
         chat_ctx=chat_ctx,
     )
     agent.start(ctx.room, participant)
+    agent.generate_reply()
 
     @agent.on("agent_speech_committed")
     @agent.on("agent_speech_interrupted")

examples/voice-pipeline-agent/gemini_voice_agent.py

@@ -1,4 +1,5 @@
 import logging
+from typing import Annotated
 
 from dotenv import load_dotenv
 from livekit.agents import (
@@ -51,6 +52,17 @@ async def entrypoint(ctx: JobContext):
     participant = await ctx.wait_for_participant()
     logger.info(f"starting voice assistant for participant {participant.identity}")
 
+    fnc_ctx = llm.FunctionContext()
+
+    @fnc_ctx.ai_callable()
+    async def get_weather(
+        location: Annotated[
+            str, llm.TypeInfo(description="The location to get the weather for")
+        ],
+    ):
+        """Called when the user asks about the weather. This function will return the weather for the given location."""
+        return f"The weather in {location} is sunny."
+
     agent = VoicePipelineAgent(
         vad=ctx.proc.userdata["vad"],
         stt=google.STT(),
@@ -59,6 +71,7 @@ async def entrypoint(ctx: JobContext):
             voice_name="en-US-Journey-D",
         ),
         chat_ctx=initial_ctx,
+        fnc_ctx=fnc_ctx,
     )
 
     agent.start(ctx.room, participant)