feat: apply latest spec changes

naomi-lgbt · naomi-lgbt · commit 4ad9eb75ae74 · 2025-04-30T11:59:17.000-07:00
diff --git a/deepgram/__init__.py b/deepgram/__init__.py
@@ -324,7 +324,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/client.py b/deepgram/client.py
@@ -338,7 +338,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/clients/__init__.py b/deepgram/clients/__init__.py
@@ -347,7 +347,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/clients/agent/__init__.py b/deepgram/clients/agent/__init__.py
@@ -22,7 +22,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/clients/agent/client.py b/deepgram/clients/agent/client.py
@@ -21,7 +21,6 @@
     ConversationTextResponse as LatestConversationTextResponse,
     UserStartedSpeakingResponse as LatestUserStartedSpeakingResponse,
     AgentThinkingResponse as LatestAgentThinkingResponse,
-    FunctionCalling as LatestFunctionCalling,
     FunctionCallRequest as LatestFunctionCallRequest,
     AgentStartedSpeakingResponse as LatestAgentStartedSpeakingResponse,
     AgentAudioDoneResponse as LatestAgentAudioDoneResponse,
@@ -72,7 +71,6 @@
 ConversationTextResponse = LatestConversationTextResponse
 UserStartedSpeakingResponse = LatestUserStartedSpeakingResponse
 AgentThinkingResponse = LatestAgentThinkingResponse
-FunctionCalling = LatestFunctionCalling
 FunctionCallRequest = LatestFunctionCallRequest
 AgentStartedSpeakingResponse = LatestAgentStartedSpeakingResponse
 AgentAudioDoneResponse = LatestAgentAudioDoneResponse
diff --git a/deepgram/clients/agent/enums.py b/deepgram/clients/agent/enums.py
@@ -21,7 +21,6 @@ class AgentWebSocketEvents(StrEnum):
     ConversationText: str = "ConversationText"
     UserStartedSpeaking: str = "UserStartedSpeaking"
     AgentThinking: str = "AgentThinking"
-    FunctionCalling: str = "FunctionCalling"
     FunctionCallRequest: str = "FunctionCallRequest"
     AgentStartedSpeaking: str = "AgentStartedSpeaking"
     AgentAudioDone: str = "AgentAudioDone"
diff --git a/deepgram/clients/agent/v1/__init__.py b/deepgram/clients/agent/v1/__init__.py
@@ -26,7 +26,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/clients/agent/v1/websocket/__init__.py b/deepgram/clients/agent/v1/websocket/__init__.py
@@ -18,7 +18,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
diff --git a/deepgram/clients/agent/v1/websocket/async_client.py b/deepgram/clients/agent/v1/websocket/async_client.py
@@ -21,7 +21,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
@@ -432,16 +431,6 @@ async def _process_text(self, message: str) -> None:
                         agent_thinking=agent_thinking_result,
                         **dict(cast(Dict[Any, Any], self._kwargs)),
                     )
-                case AgentWebSocketEvents.FunctionCalling:
-                    function_calling_result: FunctionCalling = (
-                        FunctionCalling.from_json(message)
-                    )
-                    self._logger.verbose("FunctionCalling: %s", function_calling_result)
-                    await self._emit(
-                        AgentWebSocketEvents(AgentWebSocketEvents.FunctionCalling),
-                        function_calling=function_calling_result,
-                        **dict(cast(Dict[Any, Any], self._kwargs)),
-                    )
                 case AgentWebSocketEvents.FunctionCallRequest:
                     function_call_request_result: FunctionCallRequest = (
                         FunctionCallRequest.from_json(message)
diff --git a/deepgram/clients/agent/v1/websocket/client.py b/deepgram/clients/agent/v1/websocket/client.py
@@ -21,7 +21,6 @@
     ConversationTextResponse,
     UserStartedSpeakingResponse,
     AgentThinkingResponse,
-    FunctionCalling,
     FunctionCallRequest,
     AgentStartedSpeakingResponse,
     AgentAudioDoneResponse,
@@ -427,16 +426,6 @@ def _process_text(self, message: str) -> None:
                         agent_thinking=agent_thinking_result,
                         **dict(cast(Dict[Any, Any], self._kwargs)),
                     )
-                case AgentWebSocketEvents.FunctionCalling:
-                    function_calling_result: FunctionCalling = (
-                        FunctionCalling.from_json(message)
-                    )
-                    self._logger.verbose("FunctionCalling: %s", function_calling_result)
-                    self._emit(
-                        AgentWebSocketEvents(AgentWebSocketEvents.FunctionCalling),
-                        function_calling=function_calling_result,
-                        **dict(cast(Dict[Any, Any], self._kwargs)),
-                    )
                 case AgentWebSocketEvents.FunctionCallRequest:
                     function_call_request_result: FunctionCallRequest = (
                         FunctionCallRequest.from_json(message)
diff --git a/deepgram/clients/agent/v1/websocket/options.py b/deepgram/clients/agent/v1/websocket/options.py
@@ -147,8 +147,8 @@ class ListenProvider(BaseResponse):
     This class defines the provider for the Listen model.
     """
 
-    type: str = field(default="deepgram")
-    model: str = field(default="nova-3")
+    type: str = field(default="")
+    model: str = field(default="")
     keyterms: Optional[List[str]] = field(
         default=None, metadata=dataclass_config(exclude=lambda f: f is None)
     )
@@ -427,7 +427,7 @@ class UpdateSpeakOptions(BaseResponse):
     """
 
     type: str = str(AgentWebSocketEvents.UpdateSpeak)
-    model: str = field(default="")
+    speak: Speak = field(default_factory=Speak)
 
 
 # InjectAgentMessage
diff --git a/deepgram/clients/agent/v1/websocket/response.py b/deepgram/clients/agent/v1/websocket/response.py
@@ -61,21 +61,13 @@ class UserStartedSpeakingResponse(BaseResponse):
 class AgentThinkingResponse(BaseResponse):
     """
     The server will send an AgentThinking message to inform the client of a non-verbalized agent thought.
+    You will ONLY receive this message if you have set `experimental` to true.
     """
 
     type: str
     content: str
 
 
-@dataclass
-class FunctionCalling(BaseResponse):
-    """
-    The server will sometimes send FunctionCalling messages when making function calls to help the client developer debug function calling workflows.
-    """
-
-    type: str
-
-
 @dataclass
 class FunctionCallRequest(BaseResponse):
     """
diff --git a/examples/agent/async_simple/main.py b/examples/agent/async_simple/main.py
@@ -73,9 +73,6 @@ async def on_user_started_speaking(self, user_started_speaking, **kwargs):
         async def on_agent_thinking(self, agent_thinking, **kwargs):
             print(f"\n\n{agent_thinking}\n\n")
 
-        async def on_function_calling(self, function_calling, **kwargs):
-            print(f"\n\n{function_calling}\n\n")
-
         async def on_agent_started_speaking(self, agent_started_speaking, **kwargs):
             print(f"\n\n{agent_started_speaking}\n\n")
 
@@ -100,7 +97,6 @@ async def on_unhandled(self, unhandled, **kwargs):
             AgentWebSocketEvents.UserStartedSpeaking, on_user_started_speaking
         )
         dg_connection.on(AgentWebSocketEvents.AgentThinking, on_agent_thinking)
-        dg_connection.on(AgentWebSocketEvents.FunctionCalling, on_function_calling)
         dg_connection.on(
             AgentWebSocketEvents.AgentStartedSpeaking, on_agent_started_speaking
         )
@@ -117,6 +113,7 @@ async def on_unhandled(self, unhandled, **kwargs):
         options.greeting = "Hello, this is a text to speech example using Deepgram."
         options.agent.listen.provider.keyterms = ["hello", "goodbye"]
         options.agent.listen.provider.model = "nova-3"
+        options.agent.listen.provider.type = "deepgram"
         options.language = "en"
 
 
diff --git a/examples/agent/simple/main.py b/examples/agent/simple/main.py
@@ -9,7 +9,6 @@
     DeepgramClientOptions,
     AgentWebSocketEvents,
     SettingsOptions,
-    FunctionCalling,
     FunctionCallRequest,
     FunctionCallResponse,
 )
@@ -82,9 +81,6 @@ def on_user_started_speaking(self, user_started_speaking, **kwargs):
         def on_agent_thinking(self, agent_thinking, **kwargs):
             print(f"\n\n{agent_thinking}\n\n")
 
-        def on_function_calling(self, function_calling: FunctionCalling, **kwargs):
-            print(f"\n\nFunction Calling Debug: {function_calling}\n\n")
-
         def on_function_call_request(
             self, function_call_request: FunctionCallRequest, **kwargs
         ):
@@ -122,7 +118,6 @@ def on_unhandled(self, unhandled, **kwargs):
             AgentWebSocketEvents.UserStartedSpeaking, on_user_started_speaking
         )
         dg_connection.on(AgentWebSocketEvents.AgentThinking, on_agent_thinking)
-        dg_connection.on(AgentWebSocketEvents.FunctionCalling, on_function_calling)
         dg_connection.on(
             AgentWebSocketEvents.FunctionCallRequest, on_function_call_request
         )
@@ -142,6 +137,7 @@ def on_unhandled(self, unhandled, **kwargs):
         options.greeting = "Hello, this is a text to speech example using Deepgram."
         options.agent.listen.provider.keyterms = ["hello", "goodbye"]
         options.agent.listen.provider.model = "nova-3"
+        options.agent.listen.provider.type = "deepgram"
         options.language = "en"
         if dg_connection.start(options) is False:
             print("Failed to start connection")
