livekit-plugins-google 1.2.15__tar.gz → 1.2.17__tar.gz

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.2.15
+Version: 1.2.17
 Summary: Agent Framework plugin for services from Google Cloud
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
 Project-URL: Source, https://github.com/livekit/agents
 Author: LiveKit
 License-Expression: Apache-2.0
-Keywords: audio,livekit,realtime,video,webrtc
+Keywords: ai,audio,gemini,google,livekit,realtime,video,voice
 Classifier: Intended Audience :: Developers
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python :: 3
@@ -22,7 +22,7 @@ Requires-Dist: google-auth<3,>=2
 Requires-Dist: google-cloud-speech<3,>=2
 Requires-Dist: google-cloud-texttospeech<3,>=2.27
 Requires-Dist: google-genai>=v1.23.0
-Requires-Dist: livekit-agents>=1.2.15
+Requires-Dist: livekit-agents>=1.2.17
 Description-Content-Type: text/markdown
 
 # Google AI plugin for LiveKit Agents

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/livekit/plugins/google/llm.py

@@ -136,6 +136,10 @@ class LLM(llm.LLM):
                 _, gcp_project = default_async(  # type: ignore
                     scopes=["https://www.googleapis.com/auth/cloud-platform"]
                 )
+            if not gcp_project or not gcp_location:
+                raise ValueError(
+                    "Project is required for VertexAI via project kwarg or GOOGLE_CLOUD_PROJECT environment variable"  # noqa: E501
+                )
             gemini_api_key = None  # VertexAI does not require an API key
 
         else:

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/livekit/plugins/google/realtime/api_proto.py

@@ -10,6 +10,7 @@ LiveAPIModels = Literal[
     # models supported on Gemini API
     "gemini-2.0-flash-live-001",
     "gemini-live-2.5-flash-preview",
+    "gemini-live-2.5-flash-preview-native-audio-09-2025",
     "gemini-2.5-flash-native-audio-preview-09-2025",
     "gemini-2.5-flash-preview-native-audio-dialog",
     "gemini-2.5-flash-exp-native-audio-thinking-dialog",

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/livekit/plugins/google/realtime/realtime_api.py

@@ -10,6 +10,7 @@ from collections.abc import Iterator
 from dataclasses import dataclass, field
 from typing import Literal
 
+from google.auth._default_async import default_async
 from google.genai import Client as GenAIClient, types
 from google.genai.live import AsyncSession
 from livekit import rtc
@@ -79,6 +80,7 @@ class _RealtimeOptions:
     tool_behavior: NotGivenOr[types.Behavior] = NOT_GIVEN
     tool_response_scheduling: NotGivenOr[types.FunctionResponseScheduling] = NOT_GIVEN
     thinking_config: NotGivenOr[types.ThinkingConfig] = NOT_GIVEN
+    session_resumption: NotGivenOr[types.SessionResumptionConfig] = NOT_GIVEN
 
 
 @dataclass
@@ -141,6 +143,7 @@ class RealtimeModel(llm.RealtimeModel):
         context_window_compression: NotGivenOr[types.ContextWindowCompressionConfig] = NOT_GIVEN,
         tool_behavior: NotGivenOr[types.Behavior] = NOT_GIVEN,
         tool_response_scheduling: NotGivenOr[types.FunctionResponseScheduling] = NOT_GIVEN,
+        session_resumption: NotGivenOr[types.SessionResumptionConfig] = NOT_GIVEN,
         api_version: NotGivenOr[str] = NOT_GIVEN,
         conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS,
         http_options: NotGivenOr[types.HttpOptions] = NOT_GIVEN,
@@ -182,6 +185,7 @@ class RealtimeModel(llm.RealtimeModel):
             context_window_compression (ContextWindowCompressionConfig, optional): The configuration for context window compression. Defaults to None.
             tool_behavior (Behavior, optional): The behavior for tool call. Default behavior is BLOCK in Gemini Realtime API.
             tool_response_scheduling (FunctionResponseScheduling, optional): The scheduling for tool response. Default scheduling is WHEN_IDLE.
+            session_resumption (SessionResumptionConfig, optional): The configuration for session resumption. Defaults to None.
             thinking_config (ThinkingConfig, optional): Native audio thinking configuration.
             conn_options (APIConnectOptions, optional): The configuration for the API connection. Defaults to DEFAULT_API_CONNECT_OPTIONS.
             _gemini_tools (list[LLMTool], optional): Gemini-specific tools to use for the session. This parameter is experimental and may change.
@@ -235,6 +239,10 @@ class RealtimeModel(llm.RealtimeModel):
         )
 
         if use_vertexai:
+            if not gcp_project:
+                _, gcp_project = default_async(  # type: ignore
+                    scopes=["https://www.googleapis.com/auth/cloud-platform"]
+                )
             if not gcp_project or not gcp_location:
                 raise ValueError(
                     "Project is required for VertexAI via project kwarg or GOOGLE_CLOUD_PROJECT environment variable"  # noqa: E501
@@ -278,6 +286,7 @@ class RealtimeModel(llm.RealtimeModel):
             conn_options=conn_options,
             http_options=http_options,
             thinking_config=thinking_config,
+            session_resumption=session_resumption,
         )
 
         self._sessions = weakref.WeakSet[RealtimeSession]()
@@ -382,7 +391,12 @@ class RealtimeSession(llm.RealtimeSession):
         self._response_created_futures: dict[str, asyncio.Future[llm.GenerationCreatedEvent]] = {}
         self._pending_generation_fut: asyncio.Future[llm.GenerationCreatedEvent] | None = None
 
-        self._session_resumption_handle: str | None = None
+        self._session_resumption_handle: str | None = (
+            self._opts.session_resumption.handle
+            if is_given(self._opts.session_resumption)
+            else None
+        )
+
         self._in_user_activity = False
         self._session_lock = asyncio.Lock()
         self._num_retries = 0
@@ -510,6 +524,10 @@ class RealtimeSession(llm.RealtimeSession):
             return True
         return False
 
+    @property
+    def session_resumption_handle(self) -> str | None:
+        return self._session_resumption_handle
+
     def push_audio(self, frame: rtc.AudioFrame) -> None:
         for f in self._resample_audio(frame):
             for nf in self._bstream.write(f.data.tobytes()):

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.2.15"
+__version__ = "1.2.17"

{livekit_plugins_google-1.2.15 → livekit_plugins_google-1.2.17}/pyproject.toml

@@ -10,7 +10,7 @@ readme = "README.md"
 license = "Apache-2.0"
 requires-python = ">=3.9.0"
 authors = [{ name = "LiveKit" }]
-keywords = ["webrtc", "realtime", "audio", "video", "livekit"]
+keywords = ["voice", "ai", "realtime", "audio", "video", "livekit", "google", "gemini"]
 classifiers = [
     "Intended Audience :: Developers",
     "License :: OSI Approved :: Apache Software License",
@@ -27,7 +27,7 @@ dependencies = [
     "google-cloud-speech >= 2, < 3",
     "google-cloud-texttospeech >= 2.27, < 3",
     "google-genai >= v1.23.0",
-    "livekit-agents>=1.2.15",
+    "livekit-agents>=1.2.17",
 ]
 
 [project.urls]