livekit-plugins-google 1.0.22__py3-none-any.whl → 1.0.23__py3-none-any.whl

livekit/plugins/google/beta/realtime/api_proto.py

@@ -5,9 +5,15 @@ from typing import Literal, Union
 
 from google.genai import types
 
-LiveAPIModels = Literal["gemini-2.0-flash-exp", "gemini-2.0-flash-live-001"]
+LiveAPIModels = Literal[
+    "gemini-2.0-flash-exp",
+    # models supported on Gemini API
+    "gemini-2.0-flash-live-001",
+    "gemini-2.5-flash-preview-native-audio-dialog",
+    "gemini-2.5-flash-exp-native-audio-thinking-dialog",
+]
 
-Voice = Literal["Puck", "Charon", "Kore", "Fenrir", "Aoede"]
+Voice = Literal["Puck", "Charon", "Kore", "Fenrir", "Aoede", "Leda", "Oru", "Zephyr"]
 
 
 ClientEvents = Union[

livekit/plugins/google/beta/realtime/realtime_api.py

@@ -13,7 +13,6 @@ from google import genai
 from google.genai.live import AsyncSession
 from google.genai.types import (
     AudioTranscriptionConfig,
-    AutomaticActivityDetection,
     Blob,
     Content,
     FunctionDeclaration,
@@ -86,6 +85,9 @@ class _RealtimeOptions:
     input_audio_transcription: AudioTranscriptionConfig | None
     output_audio_transcription: AudioTranscriptionConfig | None
     image_encode_options: NotGivenOr[images.EncodeOptions]
+    enable_affective_dialog: NotGivenOr[bool] = NOT_GIVEN
+    proactivity: NotGivenOr[bool] = NOT_GIVEN
+    realtime_input_config: NotGivenOr[RealtimeInputConfig] = NOT_GIVEN
 
 
 @dataclass
@@ -131,6 +133,9 @@ class RealtimeModel(llm.RealtimeModel):
         input_audio_transcription: NotGivenOr[AudioTranscriptionConfig | None] = NOT_GIVEN,
         output_audio_transcription: NotGivenOr[AudioTranscriptionConfig | None] = NOT_GIVEN,
         image_encode_options: NotGivenOr[images.EncodeOptions] = NOT_GIVEN,
+        enable_affective_dialog: NotGivenOr[bool] = NOT_GIVEN,
+        proactivity: NotGivenOr[bool] = NOT_GIVEN,
+        realtime_input_config: NotGivenOr[RealtimeInputConfig] = NOT_GIVEN,
     ) -> None:
         """
         Initializes a RealtimeModel instance for interacting with Google's Realtime API.
@@ -161,6 +166,9 @@ class RealtimeModel(llm.RealtimeModel):
             input_audio_transcription (AudioTranscriptionConfig | None, optional): The configuration for input audio transcription. Defaults to None.)
             output_audio_transcription (AudioTranscriptionConfig | None, optional): The configuration for output audio transcription. Defaults to AudioTranscriptionConfig().
             image_encode_options (images.EncodeOptions, optional): The configuration for image encoding. Defaults to DEFAULT_ENCODE_OPTIONS.
+            enable_affective_dialog (bool, optional): Whether to enable affective dialog. Defaults to False.
+            proactivity (bool, optional): Whether to enable proactive audio. Defaults to False.
+            realtime_input_config (RealtimeInputConfig, optional): The configuration for realtime input. Defaults to None.
 
         Raises:
             ValueError: If the API key is required but not found.
@@ -232,6 +240,9 @@ class RealtimeModel(llm.RealtimeModel):
             output_audio_transcription=output_audio_transcription,
             language=language,
             image_encode_options=image_encode_options,
+            enable_affective_dialog=enable_affective_dialog,
+            proactivity=proactivity,
+            realtime_input_config=realtime_input_config,
         )
 
         self._sessions = weakref.WeakSet[RealtimeSession]()
@@ -583,7 +594,7 @@ class RealtimeSession(llm.RealtimeSession):
     def _build_connect_config(self) -> LiveConnectConfig:
         temp = self._opts.temperature if is_given(self._opts.temperature) else None
 
-        return LiveConnectConfig(
+        conf = LiveConnectConfig(
             response_modalities=self._opts.response_modalities
             if is_given(self._opts.response_modalities)
             else [Modality.AUDIO],
@@ -615,11 +626,18 @@ class RealtimeSession(llm.RealtimeSession):
             input_audio_transcription=self._opts.input_audio_transcription,
             output_audio_transcription=self._opts.output_audio_transcription,
             session_resumption=SessionResumptionConfig(handle=self._session_resumption_handle),
-            realtime_input_config=RealtimeInputConfig(
-                automatic_activity_detection=AutomaticActivityDetection(),
-            ),
+            realtime_input_config=self._opts.realtime_input_config,
         )
 
+        if is_given(self._opts.proactivity):
+            conf.proactivity = {"proactive_audio": self._opts.proactivity}
+        if is_given(self._opts.enable_affective_dialog):
+            conf.enable_affective_dialog = self._opts.enable_affective_dialog
+        if is_given(self._opts.realtime_input_config):
+            conf.realtime_input_config = self._opts.realtime_input_config
+
+        return conf
+
     def _start_new_generation(self):
         if self._current_generation and not self._current_generation._done:
             logger.warning("starting new generation while another is active. Finalizing previous.")
@@ -789,6 +807,9 @@ class RealtimeSession(llm.RealtimeSession):
                 return token_details_map
 
             for token_detail in token_details:
+                if not token_detail.token_count:
+                    continue
+
                 if token_detail.modality == Modality.AUDIO:
                     token_details_map["audio_tokens"] += token_detail.token_count
                 elif token_detail.modality == Modality.TEXT:

livekit/plugins/google/llm.py

@@ -304,11 +304,8 @@ class LLMStream(llm.LLMStream):
                     or not response.candidates[0].content
                     or not response.candidates[0].content.parts
                 ):
-                    raise APIStatusError(
-                        "No candidates in the response",
-                        retryable=True,
-                        request_id=request_id,
-                    )
+                    logger.warning(f"no candidates in the response: {response}")
+                    continue
 
                 if len(response.candidates) > 1:
                     logger.warning(

livekit/plugins/google/models.py

@@ -97,6 +97,7 @@ Gender = Literal["male", "female", "neutral"]
 ChatModels = Literal[
     "gemini-2.5-pro-preview-05-06",
     "gemini-2.5-flash-preview-04-17",
+    "gemini-2.5-flash-preview-05-20",
     "gemini-2.0-flash-001",
     "gemini-2.0-flash-lite-preview-02-05",
     "gemini-2.0-pro-exp-02-05",

livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.0.22"
+__version__ = "1.0.23"

{livekit_plugins_google-1.0.22.dist-info → livekit_plugins_google-1.0.23.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.0.22
+Version: 1.0.23
 Summary: Agent Framework plugin for services from Google Cloud
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
@@ -21,8 +21,8 @@ Requires-Python: >=3.9.0
 Requires-Dist: google-auth<3,>=2
 Requires-Dist: google-cloud-speech<3,>=2
 Requires-Dist: google-cloud-texttospeech<3,>=2.24
-Requires-Dist: google-genai>=1.14.0
-Requires-Dist: livekit-agents>=1.0.22
+Requires-Dist: google-genai>=v1.16.1
+Requires-Dist: livekit-agents>=1.0.23
 Description-Content-Type: text/markdown
 
 # Google AI plugin for LiveKit Agents

{livekit_plugins_google-1.0.22.dist-info → livekit_plugins_google-1.0.23.dist-info}/RECORD

@@ -1,16 +1,16 @@
 livekit/plugins/google/__init__.py,sha256=xain2qUzU-YWhYWsLBkW8Q-szV-htpnzHTqymMPo-j0,1364
-livekit/plugins/google/llm.py,sha256=Kr9qeBZ5Dd0WCCBR_-gM3WWsVRZPCSteK8NpBsg2C5Y,16304
+livekit/plugins/google/llm.py,sha256=E1T_7cugMVN13dyAbXHVS5sC1lxRPNUemwJdV29-CPk,16206
 livekit/plugins/google/log.py,sha256=GI3YWN5YzrafnUccljzPRS_ZALkMNk1i21IRnTl2vNA,69
-livekit/plugins/google/models.py,sha256=maGlEM3hK4-5hMnH9UQMJewA7BZMrnStsFLBNoNVySg,1531
+livekit/plugins/google/models.py,sha256=hOpfbN_qdQ1ZTpCN9m9dvG2eb6WgQ3KE3WRpIeeM_T0,1569
 livekit/plugins/google/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 livekit/plugins/google/stt.py,sha256=2jk-1fHiBT8UW_n3CZsIEdMp2iBnUAlTnmefdUd8rAM,23620
 livekit/plugins/google/tts.py,sha256=FfhNfGtW8drmYDDfLLZDjaIp2GvNiIdoovgtZq4t_l8,14211
 livekit/plugins/google/utils.py,sha256=UBAbddYk7G8Nojg6bSC7_xN2pdl9qhs86HGhKYFuf9M,10509
-livekit/plugins/google/version.py,sha256=-8dkOE2vDSF9WN8VoBrSwU2sb5YBGFuwPnSQXQ-uaYM,601
+livekit/plugins/google/version.py,sha256=BRUqwxRBnPVqEcIODJdaZHGAanu4zkwM4NsAQjNtUEM,601
 livekit/plugins/google/beta/__init__.py,sha256=5PnoG3Ux24bjzMSzmTeSVljE9EINivGcbWUEV6egGnM,216
 livekit/plugins/google/beta/realtime/__init__.py,sha256=_fW2NMN22F-hnQ4xAJ_g5lPbR7CvM_xXzSWlUQY-E-U,188
-livekit/plugins/google/beta/realtime/api_proto.py,sha256=Fyrejs3SG0EjOPCCFLEnWXKEUxCff47PMWk2VsKJm5E,594
-livekit/plugins/google/beta/realtime/realtime_api.py,sha256=yYB5fKXl_aaMH_ZSpfUlfOTUg4eRqqRENLTZhZMfBMc,36253
-livekit_plugins_google-1.0.22.dist-info/METADATA,sha256=S4bQZr4NhWrAI6vyJi299sh5lsD5eVMNfxvN9__xAMY,1908
-livekit_plugins_google-1.0.22.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_google-1.0.22.dist-info/RECORD,,
+livekit/plugins/google/beta/realtime/api_proto.py,sha256=NfE7xr2N3JOu7gVfWbAmDcEhs8vuZgMRu5vpScPJzsg,776
+livekit/plugins/google/beta/realtime/realtime_api.py,sha256=fgN2InMMCQL8JAHm-6J-SekzS5ymeH-hMRLzSW86Qkw,37477
+livekit_plugins_google-1.0.23.dist-info/METADATA,sha256=69J1PJEwdaM6jWeMUXpbaU8A0quqi3UjDb5884qG9mI,1909
+livekit_plugins_google-1.0.23.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_google-1.0.23.dist-info/RECORD,,