Include examples in type checking

Remove `examples/` from the `pyrightconfig.json` ignore list and fix
the resulting type errors across all example files. Common fixes:

- Required API keys: `os.getenv("X")` -> `os.environ["X"]` so the
  return type is `str` rather than `str | None`, and misconfiguration
  fails fast.
- Narrow `LLMContextMessage` union members with `isinstance(..., dict)`
  before dict-style access.
- `assert isinstance(params.llm, ...)` before calling service-specific
  methods that aren't on the base `LLMService`.
- Guard optional frame fields (e.g. `LLMSearchResponseFrame.search_result`)
  before use.

examples/function-calling/function-calling-anthropic-async-stream.py

@@ -107,17 +107,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = AnthropicLLMService(
-        api_key=os.getenv("ANTHROPIC_API_KEY"),
+        api_key=os.environ["ANTHROPIC_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=AnthropicLLMService.Settings(
             system_instruction=(

examples/function-calling/function-calling-anthropic-async.py

@@ -66,17 +66,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = AnthropicLLMService(
-        api_key=os.getenv("ANTHROPIC_API_KEY"),
+        api_key=os.environ["ANTHROPIC_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=AnthropicLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-anthropic-video.py

@@ -86,10 +86,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
@@ -97,7 +97,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     # Anthropic for vision analysis
     llm = AnthropicLLMService(
-        api_key=os.getenv("ANTHROPIC_API_KEY"),
+        api_key=os.environ["ANTHROPIC_API_KEY"],
         settings=AnthropicLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-anthropic.py

@@ -65,17 +65,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = AnthropicLLMService(
-        api_key=os.getenv("ANTHROPIC_API_KEY"),
+        api_key=os.environ["ANTHROPIC_API_KEY"],
         settings=AnthropicLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-aws-video.py

@@ -86,10 +86,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),

examples/function-calling/function-calling-azure.py

@@ -60,18 +60,18 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = AzureLLMService(
-        api_key=os.getenv("AZURE_CHATGPT_API_KEY"),
-        endpoint=os.getenv("AZURE_CHATGPT_ENDPOINT"),
+        api_key=os.environ["AZURE_CHATGPT_API_KEY"],
+        endpoint=os.environ["AZURE_CHATGPT_ENDPOINT"],
         settings=AzureLLMService.Settings(
             model=os.getenv("AZURE_CHATGPT_MODEL"),
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-cerebras.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = CerebrasLLMService(
-        api_key=os.getenv("CEREBRAS_API_KEY"),
+        api_key=os.environ["CEREBRAS_API_KEY"],
         settings=CerebrasLLMService.Settings(
             system_instruction="""You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.
 

examples/function-calling/function-calling-deepseek.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = DeepSeekLLMService(
-        api_key=os.getenv("DEEPSEEK_API_KEY"),
+        api_key=os.environ["DEEPSEEK_API_KEY"],
         settings=DeepSeekLLMService.Settings(
             model="deepseek-chat",
             system_instruction="""You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.

examples/function-calling/function-calling-direct.py

@@ -76,17 +76,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAILLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-fireworks.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = FireworksLLMService(
-        api_key=os.getenv("FIREWORKS_API_KEY"),
+        api_key=os.environ["FIREWORKS_API_KEY"],
         settings=FireworksLLMService.Settings(
             model="accounts/fireworks/models/gpt-oss-20b",
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-google-async-stream.py

@@ -107,17 +107,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = GoogleLLMService(
-        api_key=os.getenv("GOOGLE_API_KEY"),
+        api_key=os.environ["GOOGLE_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=GoogleLLMService.Settings(
             system_instruction=(

examples/function-calling/function-calling-google-async.py

@@ -98,10 +98,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
@@ -127,7 +127,7 @@ indicate you should use the get_image tool are:
 """
 
     llm = GoogleLLMService(
-        api_key=os.getenv("GOOGLE_API_KEY"),
+        api_key=os.environ["GOOGLE_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=GoogleLLMService.Settings(
             system_instruction=system_prompt,

examples/function-calling/function-calling-google-vertex.py

@@ -60,19 +60,19 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = ElevenLabsTTSService(
-        api_key=os.getenv("ELEVENLABS_API_KEY", ""),
+        api_key=os.environ["ELEVENLABS_API_KEY"],
         settings=ElevenLabsTTSService.Settings(
-            voice=os.getenv("ELEVENLABS_VOICE_ID", ""),
+            voice=os.getenv("ELEVENLABS_VOICE_ID", "Xb7hH8MSUJpSbSDYk0k2"),
         ),
     )
 
     llm = GoogleVertexLLMService(
-        credentials=os.getenv("GOOGLE_VERTEX_TEST_CREDENTIALS"),
-        project_id=os.getenv("GOOGLE_CLOUD_PROJECT_ID"),
-        location=os.getenv("GOOGLE_CLOUD_LOCATION"),
+        credentials=os.environ["GOOGLE_VERTEX_TEST_CREDENTIALS"],
+        project_id=os.environ["GOOGLE_CLOUD_PROJECT_ID"],
+        location=os.environ["GOOGLE_CLOUD_LOCATION"],
         settings=GoogleVertexLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),
@@ -103,14 +103,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
     tools = ToolsSchema(standard_tools=[weather_function])
 
-    messages = [
-        {
-            "role": "developer",
-            "content": "Start a conversation with 'Hey there' to get the current weather.",
-        },
-    ]
-
-    context = LLMContext(messages, tools)
+    context = LLMContext(tools=tools)
     user_aggregator, assistant_aggregator = LLMContextAggregatorPair(
         context,
         user_params=LLMUserAggregatorParams(vad_analyzer=SileroVADAnalyzer()),
@@ -141,6 +134,12 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     async def on_client_connected(transport, client):
         logger.info(f"Client connected")
         # Kick off the conversation.
+        context.add_message(
+            {
+                "role": "developer",
+                "content": "Please introduce yourself to the user.",
+            }
+        )
         await task.queue_frames([LLMRunFrame()])
 
     @transport.event_handler("on_client_disconnected")

examples/function-calling/function-calling-google-video.py

@@ -86,10 +86,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
@@ -97,7 +97,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
 
     # Google Gemini model for vision analysis
     llm = GoogleLLMService(
-        api_key=os.getenv("GOOGLE_API_KEY"),
+        api_key=os.environ["GOOGLE_API_KEY"],
         settings=GoogleLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-google.py

@@ -96,10 +96,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
@@ -125,7 +125,7 @@ indicate you should use the get_image tool are:
 """
 
     llm = GoogleLLMService(
-        api_key=os.getenv("GOOGLE_API_KEY"),
+        api_key=os.environ["GOOGLE_API_KEY"],
         settings=GoogleLLMService.Settings(
             system_instruction=system_prompt,
         ),

examples/function-calling/function-calling-grok.py

@@ -62,10 +62,10 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
     async with aiohttp.ClientSession() as session:
-        stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+        stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
         tts = XAIHttpTTSService(
-            api_key=os.getenv("XAI_API_KEY"),
+            api_key=os.environ["XAI_API_KEY"],
             aiohttp_session=session,
             settings=XAIHttpTTSService.Settings(
                 voice="eve",
@@ -73,7 +73,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
         )
 
         llm = GrokLLMService(
-            api_key=os.getenv("XAI_API_KEY"),
+            api_key=os.environ["XAI_API_KEY"],
             settings=GrokLLMService.Settings(
                 system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
             ),

examples/function-calling/function-calling-groq.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = GroqSTTService(api_key=os.getenv("GROQ_API_KEY"))
+    stt = GroqSTTService(api_key=os.environ["GROQ_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = GroqLLMService(
-        api_key=os.getenv("GROQ_API_KEY"),
+        api_key=os.environ["GROQ_API_KEY"],
         settings=GroqLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-mistral.py

@@ -63,17 +63,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = MistralLLMService(
-        api_key=os.getenv("MISTRAL_API_KEY"),
+        api_key=os.environ["MISTRAL_API_KEY"],
         settings=MistralLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-moondream-video.py

@@ -117,17 +117,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAILLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-nebius.py

@@ -63,17 +63,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = NebiusLLMService(
-        api_key=os.getenv("NEBIUS_API_KEY"),
+        api_key=os.environ["NEBIUS_API_KEY"],
         settings=NebiusLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-novita.py

@@ -63,17 +63,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = NovitaLLMService(
-        api_key=os.getenv("NOVITA_API_KEY"),
+        api_key=os.environ["NOVITA_API_KEY"],
         settings=NovitaLLMService.Settings(
             model="openai/gpt-oss-120b",
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-nvidia.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = NvidiaLLMService(
-        api_key=os.getenv("NVIDIA_API_KEY"),
+        api_key=os.environ["NVIDIA_API_KEY"],
         settings=NvidiaLLMService.Settings(
             model="nvidia/llama-3.3-nemotron-super-49b-v1.5",
             # Recommended when turning thinking off

examples/function-calling/function-calling-ollama.py

@@ -64,10 +64,10 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),

examples/function-calling/function-calling-openai-async-stream.py

@@ -107,17 +107,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=OpenAILLMService.Settings(
             system_instruction=(

examples/function-calling/function-calling-openai-async.py

@@ -70,7 +70,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
     stt = OpenAISTTService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAISTTService.Settings(
             model="gpt-4o-transcribe",
             prompt="Expect words related weather, such as temperature and conditions. And restaurant names.",
@@ -78,7 +78,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     tts = OpenAITTSService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAITTSService.Settings(
             voice="ballad",
         ),
@@ -86,7 +86,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=OpenAILLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-openai-responses-async-stream.py

@@ -107,17 +107,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=OpenAIResponsesLLMService.Settings(
             system_instruction=(

examples/function-calling/function-calling-openai-responses-async.py

@@ -66,17 +66,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         enable_async_tool_cancellation=True,
         settings=OpenAIResponsesLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-openai-responses-http.py

@@ -63,17 +63,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesHttpLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAIResponsesHttpLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-openai-responses-video-http.py

@@ -87,17 +87,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesHttpLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAIResponsesHttpLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-openai-responses-video.py

@@ -87,17 +87,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAIResponsesLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-openai-responses.py

@@ -63,17 +63,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAIResponsesLLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAIResponsesLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-openai-video.py

@@ -87,17 +87,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAILLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way. You are able to describe images from the user camera.",
         ),

examples/function-calling/function-calling-openai.py

@@ -64,7 +64,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
     stt = OpenAISTTService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAISTTService.Settings(
             model="gpt-4o-transcribe",
             prompt="Expect words related weather, such as temperature and conditions. And restaurant names.",
@@ -72,7 +72,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     tts = OpenAITTSService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAITTSService.Settings(
             voice="ballad",
         ),
@@ -80,7 +80,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     llm = OpenAILLMService(
-        api_key=os.getenv("OPENAI_API_KEY"),
+        api_key=os.environ["OPENAI_API_KEY"],
         settings=OpenAILLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-openrouter.py

@@ -60,11 +60,11 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = AzureTTSService(
-        api_key=os.getenv("AZURE_SPEECH_API_KEY"),
-        region=os.getenv("AZURE_SPEECH_REGION"),
+        api_key=os.environ["AZURE_SPEECH_API_KEY"],
+        region=os.environ["AZURE_SPEECH_REGION"],
         settings=AzureTTSService.Settings(
             voice="en-US-JennyNeural",
             language="en-US",
@@ -74,7 +74,7 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     )
 
     llm = OpenRouterLLMService(
-        api_key=os.getenv("OPENROUTER_API_KEY"),
+        api_key=os.environ["OPENROUTER_API_KEY"],
         settings=OpenRouterLLMService.Settings(
             model="openai/gpt-4o-2024-11-20",
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-perplexity.py

@@ -58,17 +58,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = PerplexityLLMService(
-        api_key=os.getenv("PERPLEXITY_API_KEY"),
+        api_key=os.environ["PERPLEXITY_API_KEY"],
         settings=PerplexityLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-qwen.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = QwenLLMService(
-        api_key=os.getenv("QWEN_API_KEY"),
+        api_key=os.environ["QWEN_API_KEY"],
         model="qwen2.5-72b-instruct",
         settings=QwenLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",

examples/function-calling/function-calling-sambanova.py

@@ -61,18 +61,18 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
     stt = DeepgramSTTService(
-        api_key=os.getenv("DEEPGRAM_API_KEY"),
+        api_key=os.environ["DEEPGRAM_API_KEY"],
     )
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = SambaNovaLLMService(
-        api_key=os.getenv("SAMBANOVA_API_KEY"),
+        api_key=os.environ["SAMBANOVA_API_KEY"],
         settings=SambaNovaLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-sarvam.py

@@ -64,21 +64,21 @@ async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
     stt = SarvamSTTService(
-        api_key=os.getenv("SARVAM_API_KEY"),
+        api_key=os.environ["SARVAM_API_KEY"],
         settings=SarvamSTTService.Settings(
             model="saaras:v3",
         ),
     )
 
     tts = SarvamTTSService(
-        api_key=os.getenv("SARVAM_API_KEY"),
+        api_key=os.environ["SARVAM_API_KEY"],
         settings=SarvamTTSService.Settings(
             model="bulbul:v3",
             voice="shubh",
         ),
     )
     llm = SarvamLLMService(
-        api_key=os.getenv("SARVAM_API_KEY"),
+        api_key=os.environ["SARVAM_API_KEY"],
         settings=SarvamLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),

examples/function-calling/function-calling-together.py

@@ -60,17 +60,17 @@ transport_params = {
 async def run_bot(transport: BaseTransport, runner_args: RunnerArguments):
     logger.info(f"Starting bot")
 
-    stt = DeepgramSTTService(api_key=os.getenv("DEEPGRAM_API_KEY"))
+    stt = DeepgramSTTService(api_key=os.environ["DEEPGRAM_API_KEY"])
 
     tts = CartesiaTTSService(
-        api_key=os.getenv("CARTESIA_API_KEY"),
+        api_key=os.environ["CARTESIA_API_KEY"],
         settings=CartesiaTTSService.Settings(
             voice="71a7ad14-091c-4e8e-a314-022ece01c121",  # British Reading Lady
         ),
     )
 
     llm = TogetherLLMService(
-        api_key=os.getenv("TOGETHER_API_KEY"),
+        api_key=os.environ["TOGETHER_API_KEY"],
         settings=TogetherLLMService.Settings(
             system_instruction="You are a helpful assistant in a voice conversation. Your responses will be spoken aloud, so avoid emojis, bullet points, or other formatting that can't be spoken. Respond to what the user said in a creative, helpful, and brief way.",
         ),