[VoiceLive] Release 1.2.0b1 - Add MCP support (#44013)

* ADD MCP support

* fix docs

---------

Co-authored-by: Xiting Zhang <xitzhang@microsoft.com>

Author: xitzhang

sdk/ai/azure-ai-voicelive/CHANGELOG.md

@@ -1,5 +1,18 @@
 # Release History
 
+## 1.2.0b1 (2025-11-14)
+
+### Features Added
+
+- **MCP (Model Context Protocol) Support**: Added comprehensive support for Model Context Protocol integration:
+  - `MCPServer` tool type for defining MCP server configurations with authorization, headers, and approval requirements
+  - `MCPTool` model for representing MCP tool definitions with input schemas and annotations
+  - `MCPApprovalType` enum for controlling approval workflows (`never`, `always`, or tool-specific)
+  - New item types: `MCPApprovalResponseRequestItem`, `ResponseMCPApprovalRequestItem`, `ResponseMCPApprovalResponseItem`, `ResponseMCPCallItem`, and `ResponseMCPListToolItem`
+  - New server events: `ServerEventMcpListToolsInProgress`, `ServerEventMcpListToolsCompleted`, `ServerEventMcpListToolsFailed`, `ServerEventResponseMcpCallArgumentsDelta`, and `ServerEventResponseMcpCallArgumentsDone`
+  - Client event `MCP_APPROVAL_RESPONSE` for responding to approval requests
+  - Enhanced `ItemType` enum with MCP-related types: `mcp_list_tools`, `mcp_call`, `mcp_approval_request`, and `mcp_approval_response`
+
 ## 1.1.0 (2025-11-03)
 
 ### Features Added

sdk/ai/azure-ai-voicelive/apiview-properties.json

@@ -54,6 +54,9 @@
         "azure.ai.voicelive.models.InputTextContentPart": "VoiceLive.InputTextContentPart",
         "azure.ai.voicelive.models.InputTokenDetails": "VoiceLive.InputTokenDetails",
         "azure.ai.voicelive.models.LogProbProperties": "VoiceLive.LogProbProperties",
+        "azure.ai.voicelive.models.MCPApprovalResponseRequestItem": "VoiceLive.MCPApprovalResponseRequestItem",
+        "azure.ai.voicelive.models.MCPServer": "VoiceLive.MCPServer",
+        "azure.ai.voicelive.models.MCPTool": "VoiceLive.MCPTool",
         "azure.ai.voicelive.models.OpenAIVoice": "VoiceLive.OpenAIVoice",
         "azure.ai.voicelive.models.OutputTextContentPart": "VoiceLive.OutputTextContentPart",
         "azure.ai.voicelive.models.OutputTokenDetails": "VoiceLive.OutputTokenDetails",
@@ -70,6 +73,10 @@
         "azure.ai.voicelive.models.ResponseFunctionCallItem": "VoiceLive.ResponseFunctionCallItem",
         "azure.ai.voicelive.models.ResponseFunctionCallOutputItem": "VoiceLive.ResponseFunctionCallOutputItem",
         "azure.ai.voicelive.models.ResponseIncompleteDetails": "VoiceLive.ResponseIncompleteDetails",
+        "azure.ai.voicelive.models.ResponseMCPApprovalRequestItem": "VoiceLive.ResponseMCPApprovalRequestItem",
+        "azure.ai.voicelive.models.ResponseMCPApprovalResponseItem": "VoiceLive.ResponseMCPApprovalResponseItem",
+        "azure.ai.voicelive.models.ResponseMCPCallItem": "VoiceLive.ResponseMCPCallItem",
+        "azure.ai.voicelive.models.ResponseMCPListToolItem": "VoiceLive.ResponseMCPListToolItem",
         "azure.ai.voicelive.models.ResponseMessageItem": "VoiceLive.ResponseMessageItem",
         "azure.ai.voicelive.models.ResponseSession": "VoiceLive.ResponseSession",
         "azure.ai.voicelive.models.ResponseTextContentPart": "VoiceLive.ResponseTextContentPart",
@@ -87,6 +94,9 @@
         "azure.ai.voicelive.models.ServerEventInputAudioBufferCommitted": "VoiceLive.ServerEventInputAudioBufferCommitted",
         "azure.ai.voicelive.models.ServerEventInputAudioBufferSpeechStarted": "VoiceLive.ServerEventInputAudioBufferSpeechStarted",
         "azure.ai.voicelive.models.ServerEventInputAudioBufferSpeechStopped": "VoiceLive.ServerEventInputAudioBufferSpeechStopped",
+        "azure.ai.voicelive.models.ServerEventMcpListToolsCompleted": "VoiceLive.ServerEventMcpListToolsCompleted",
+        "azure.ai.voicelive.models.ServerEventMcpListToolsFailed": "VoiceLive.ServerEventMcpListToolsFailed",
+        "azure.ai.voicelive.models.ServerEventMcpListToolsInProgress": "VoiceLive.ServerEventMcpListToolsInProgress",
         "azure.ai.voicelive.models.ServerEventResponseAnimationBlendshapeDelta": "VoiceLive.ServerEventResponseAnimationBlendshapeDelta",
         "azure.ai.voicelive.models.ServerEventResponseAnimationBlendshapeDone": "VoiceLive.ServerEventResponseAnimationBlendshapeDone",
         "azure.ai.voicelive.models.ServerEventResponseAnimationVisemeDelta": "VoiceLive.ServerEventResponseAnimationVisemeDelta",
@@ -103,6 +113,8 @@
         "azure.ai.voicelive.models.ServerEventResponseDone": "VoiceLive.ServerEventResponseDone",
         "azure.ai.voicelive.models.ServerEventResponseFunctionCallArgumentsDelta": "VoiceLive.ServerEventResponseFunctionCallArgumentsDelta",
         "azure.ai.voicelive.models.ServerEventResponseFunctionCallArgumentsDone": "VoiceLive.ServerEventResponseFunctionCallArgumentsDone",
+        "azure.ai.voicelive.models.ServerEventResponseMcpCallArgumentsDelta": "VoiceLive.ServerEventResponseMcpCallArgumentsDelta",
+        "azure.ai.voicelive.models.ServerEventResponseMcpCallArgumentsDone": "VoiceLive.ServerEventResponseMcpCallArgumentsDone",
         "azure.ai.voicelive.models.ServerEventResponseOutputItemAdded": "VoiceLive.ServerEventResponseOutputItemAdded",
         "azure.ai.voicelive.models.ServerEventResponseOutputItemDone": "VoiceLive.ServerEventResponseOutputItemDone",
         "azure.ai.voicelive.models.ServerEventResponseTextDelta": "VoiceLive.ServerEventResponseTextDelta",
@@ -132,6 +144,7 @@
         "azure.ai.voicelive.models.PersonalVoiceModels": "VoiceLive.PersonalVoiceModels",
         "azure.ai.voicelive.models.OutputAudioFormat": "VoiceLive.OutputAudioFormat",
         "azure.ai.voicelive.models.ToolType": "VoiceLive.ToolType",
+        "azure.ai.voicelive.models.MCPApprovalType": "VoiceLive.MCPApprovalType",
         "azure.ai.voicelive.models.AnimationOutputType": "VoiceLive.AnimationOutputType",
         "azure.ai.voicelive.models.InputAudioFormat": "VoiceLive.InputAudioFormat",
         "azure.ai.voicelive.models.TurnDetectionType": "VoiceLive.TurnDetectionType",

sdk/ai/azure-ai-voicelive/azure/ai/voicelive/_version.py

@@ -6,4 +6,4 @@
 # Changes may cause incorrect behavior and will be lost if the code is regenerated.
 # --------------------------------------------------------------------------
 
-VERSION = "1.1.0"
+VERSION = "1.2.0b1"

sdk/ai/azure-ai-voicelive/azure/ai/voicelive/models/__init__.py

@@ -63,6 +63,9 @@
     InputTextContentPart,
     InputTokenDetails,
     LogProbProperties,
+    MCPApprovalResponseRequestItem,
+    MCPServer,
+    MCPTool,
     MessageContentPart,
     MessageItem,
     OpenAIVoice,
@@ -80,6 +83,10 @@
     ResponseFunctionCallOutputItem,
     ResponseIncompleteDetails,
     ResponseItem,
+    ResponseMCPApprovalRequestItem,
+    ResponseMCPApprovalResponseItem,
+    ResponseMCPCallItem,
+    ResponseMCPListToolItem,
     ResponseMessageItem,
     ResponseSession,
     ResponseStatusDetails,
@@ -98,6 +105,9 @@
     ServerEventInputAudioBufferCommitted,
     ServerEventInputAudioBufferSpeechStarted,
     ServerEventInputAudioBufferSpeechStopped,
+    ServerEventMcpListToolsCompleted,
+    ServerEventMcpListToolsFailed,
+    ServerEventMcpListToolsInProgress,
     ServerEventResponseAnimationBlendshapeDelta,
     ServerEventResponseAnimationBlendshapeDone,
     ServerEventResponseAnimationVisemeDelta,
@@ -114,6 +124,8 @@
     ServerEventResponseDone,
     ServerEventResponseFunctionCallArgumentsDelta,
     ServerEventResponseFunctionCallArgumentsDone,
+    ServerEventResponseMcpCallArgumentsDelta,
+    ServerEventResponseMcpCallArgumentsDone,
     ServerEventResponseOutputItemAdded,
     ServerEventResponseOutputItemDone,
     ServerEventResponseTextDelta,
@@ -146,6 +158,7 @@
     InputAudioFormat,
     ItemParamStatus,
     ItemType,
+    MCPApprovalType,
     MessageRole,
     Modality,
     OpenAIVoiceName,
@@ -212,6 +225,9 @@
     "InputTextContentPart",
     "InputTokenDetails",
     "LogProbProperties",
+    "MCPApprovalResponseRequestItem",
+    "MCPServer",
+    "MCPTool",
     "MessageContentPart",
     "MessageItem",
     "OpenAIVoice",
@@ -229,6 +245,10 @@
     "ResponseFunctionCallOutputItem",
     "ResponseIncompleteDetails",
     "ResponseItem",
+    "ResponseMCPApprovalRequestItem",
+    "ResponseMCPApprovalResponseItem",
+    "ResponseMCPCallItem",
+    "ResponseMCPListToolItem",
     "ResponseMessageItem",
     "ResponseSession",
     "ResponseStatusDetails",
@@ -247,6 +267,9 @@
     "ServerEventInputAudioBufferCommitted",
     "ServerEventInputAudioBufferSpeechStarted",
     "ServerEventInputAudioBufferSpeechStopped",
+    "ServerEventMcpListToolsCompleted",
+    "ServerEventMcpListToolsFailed",
+    "ServerEventMcpListToolsInProgress",
     "ServerEventResponseAnimationBlendshapeDelta",
     "ServerEventResponseAnimationBlendshapeDone",
     "ServerEventResponseAnimationVisemeDelta",
@@ -263,6 +286,8 @@
     "ServerEventResponseDone",
     "ServerEventResponseFunctionCallArgumentsDelta",
     "ServerEventResponseFunctionCallArgumentsDone",
+    "ServerEventResponseMcpCallArgumentsDelta",
+    "ServerEventResponseMcpCallArgumentsDone",
     "ServerEventResponseOutputItemAdded",
     "ServerEventResponseOutputItemDone",
     "ServerEventResponseTextDelta",
@@ -292,6 +317,7 @@
     "InputAudioFormat",
     "ItemParamStatus",
     "ItemType",
+    "MCPApprovalType",
     "MessageRole",
     "Modality",
     "OpenAIVoiceName",

sdk/ai/azure-ai-voicelive/azure/ai/voicelive/models/_enums.py

@@ -56,6 +56,7 @@ class ClientEventType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
     RESPONSE_CREATE = "response.create"
     RESPONSE_CANCEL = "response.cancel"
     SESSION_AVATAR_CONNECT = "session.avatar.connect"
+    MCP_APPROVAL_RESPONSE = "mcp_approval_response"
 
 
 class ContentPartType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
@@ -106,6 +107,19 @@ class ItemType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
     MESSAGE = "message"
     FUNCTION_CALL = "function_call"
     FUNCTION_CALL_OUTPUT = "function_call_output"
+    MCP_LIST_TOOLS = "mcp_list_tools"
+    MCP_CALL = "mcp_call"
+    MCP_APPROVAL_REQUEST = "mcp_approval_request"
+    MCP_APPROVAL_RESPONSE = "mcp_approval_response"
+
+
+class MCPApprovalType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
+    """The available set of MCP approval types."""
+
+    NEVER = "never"
+    """Approval is never required."""
+    ALWAYS = "always"
+    """Approval is always required."""
 
 
 class MessageRole(str, Enum, metaclass=CaseInsensitiveEnumMeta):
@@ -235,6 +249,16 @@ class ServerEventType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
     RESPONSE_ANIMATION_VISEME_DONE = "response.animation_viseme.done"
     RESPONSE_FUNCTION_CALL_ARGUMENTS_DELTA = "response.function_call_arguments.delta"
     RESPONSE_FUNCTION_CALL_ARGUMENTS_DONE = "response.function_call_arguments.done"
+    MCP_LIST_TOOLS_IN_PROGRESS = "mcp_list_tools.in_progress"
+    MCP_LIST_TOOLS_COMPLETED = "mcp_list_tools.completed"
+    MCP_LIST_TOOLS_FAILED = "mcp_list_tools.failed"
+    RESPONSE_MCP_CALL_ARGUMENTS_DELTA = "response.mcp_call_arguments.delta"
+    RESPONSE_MCP_CALL_ARGUMENTS_DONE = "response.mcp_call_arguments.done"
+    MCP_APPROVAL_REQUEST = "mcp_approval_request"
+    MCP_APPROVAL_RESPONSE = "mcp_approval_response"
+    RESPONSE_MCP_CALL_IN_PROGRESS = "response.mcp_call.in_progress"
+    RESPONSE_MCP_CALL_COMPLETED = "response.mcp_call.completed"
+    RESPONSE_MCP_CALL_FAILED = "response.mcp_call.failed"
 
 
 class ToolChoiceLiteral(str, Enum, metaclass=CaseInsensitiveEnumMeta):
@@ -254,6 +278,7 @@ class ToolType(str, Enum, metaclass=CaseInsensitiveEnumMeta):
     """
 
     FUNCTION = "function"
+    MCP = "mcp"
 
 
 class TurnDetectionType(str, Enum, metaclass=CaseInsensitiveEnumMeta):