livekit-plugins-google 1.2.0__tar.gz → 1.2.2__tar.gz

{livekit_plugins_google-1.2.0 → livekit_plugins_google-1.2.2}/.gitignore

@@ -169,4 +169,7 @@ node_modules
 
 credentials.json
 pyrightconfig.json
-docs/
+docs/
+
+# Database files
+*.db

{livekit_plugins_google-1.2.0 → livekit_plugins_google-1.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.2.0
+Version: 1.2.2
 Summary: Agent Framework plugin for services from Google Cloud
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
@@ -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>=1.2.0
+Requires-Dist: livekit-agents>=1.2.2
 Description-Content-Type: text/markdown
 
 # Google AI plugin for LiveKit Agents

{livekit_plugins_google-1.2.0 → livekit_plugins_google-1.2.2}/livekit/plugins/google/beta/realtime/realtime_api.py

@@ -786,6 +786,10 @@ class RealtimeSession(llm.RealtimeSession):
             logger.warning("starting new generation while another is active. Finalizing previous.")
             self._mark_current_generation_done()
 
+        # emit input_speech_started event before starting a new generation
+        # to interrupt the previous audio playout if any
+        self._handle_input_speech_started()
+
         response_id = utils.shortuuid("GR_")
         self._current_generation = _ResponseGeneration(
             message_ch=utils.aio.Chan[llm.MessageGeneration](),
@@ -882,6 +886,9 @@ class RealtimeSession(llm.RealtimeSession):
         if not self._current_generation or self._current_generation._done:
             return
 
+        # emit input_speech_stopped event after the generation is done
+        self._handle_input_speech_stopped()
+
         gen = self._current_generation
 
         # The only way we'd know that the transcription is complete is by when they are
@@ -926,6 +933,14 @@ class RealtimeSession(llm.RealtimeSession):
     def _handle_input_speech_started(self) -> None:
         self.emit("input_speech_started", llm.InputSpeechStartedEvent())
 
+    def _handle_input_speech_stopped(self) -> None:
+        self.emit(
+            "input_speech_stopped",
+            llm.InputSpeechStoppedEvent(
+                user_transcription_enabled=self._realtime_model.capabilities.user_transcription
+            ),
+        )
+
     def _handle_tool_calls(self, tool_call: types.LiveServerToolCall) -> None:
         if not self._current_generation:
             logger.warning("received tool call but no active generation.")

{livekit_plugins_google-1.2.0 → livekit_plugins_google-1.2.2}/livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.2.0"
+__version__ = "1.2.2"

{livekit_plugins_google-1.2.0 → livekit_plugins_google-1.2.2}/pyproject.toml

@@ -27,7 +27,7 @@ dependencies = [
     "google-cloud-speech >= 2, < 3",
     "google-cloud-texttospeech >= 2.27, < 3",
     "google-genai >= v1.23.0",
-    "livekit-agents>=1.2.0",
+    "livekit-agents>=1.2.2",
 ]
 
 [project.urls]