livekit-plugins-google 1.0.17__py3-none-any.whl → 1.0.18__py3-none-any.whl

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

@@ -9,6 +9,7 @@ from collections.abc import Iterator
 from dataclasses import dataclass
 
 from google import genai
+from google.genai.live import AsyncSession
 from google.genai.types import (
     AudioTranscriptionConfig,
     Blob,
@@ -25,6 +26,7 @@ from google.genai.types import (
     Modality,
     Part,
     PrebuiltVoiceConfig,
+    SessionResumptionConfig,
     SpeechConfig,
     Tool,
     UsageMetadata,
@@ -62,6 +64,7 @@ class _RealtimeOptions:
     model: LiveAPIModels | str
     api_key: str | None
     voice: Voice | str
+    language: NotGivenOr[str]
     response_modalities: NotGivenOr[list[Modality]]
     vertexai: bool
     project: str | None
@@ -101,6 +104,7 @@ class RealtimeModel(llm.RealtimeModel):
         model: LiveAPIModels | str = "gemini-2.0-flash-live-001",
         api_key: NotGivenOr[str] = NOT_GIVEN,
         voice: Voice | str = "Puck",
+        language: NotGivenOr[str] = NOT_GIVEN,
         modalities: NotGivenOr[list[Modality]] = NOT_GIVEN,
         vertexai: bool = False,
         project: NotGivenOr[str] = NOT_GIVEN,
@@ -131,6 +135,7 @@ class RealtimeModel(llm.RealtimeModel):
             modalities (list[Modality], optional): Modalities to use, such as ["TEXT", "AUDIO"]. Defaults to ["AUDIO"].
             model (str, optional): The name of the model to use. Defaults to "gemini-2.0-flash-live-001".
             voice (api_proto.Voice, optional): Voice setting for audio outputs. Defaults to "Puck".
+            language (str, optional): The language(BCP-47 Code) to use for the API. supported languages - https://ai.google.dev/gemini-api/docs/live#supported-languages
             temperature (float, optional): Sampling temperature for response generation. Defaults to 0.8.
             vertexai (bool, optional): Whether to use VertexAI for the API. Defaults to False.
                 project (str, optional): The project id to use for the API. Defaults to None. (for vertexai)
@@ -195,6 +200,7 @@ class RealtimeModel(llm.RealtimeModel):
             instructions=instructions,
             input_audio_transcription=input_audio_transcription,
             output_audio_transcription=output_audio_transcription,
+            language=language,
         )
 
         self._sessions = weakref.WeakSet[RealtimeSession]()
@@ -247,12 +253,14 @@ class RealtimeSession(llm.RealtimeSession):
         self._main_atask = asyncio.create_task(self._main_task(), name="gemini-realtime-session")
 
         self._current_generation: _ResponseGeneration | None = None
-        self._active_session: genai.LiveSession | None = None
+        self._active_session: AsyncSession | None = None
         # indicates if the underlying session should end
         self._session_should_close = asyncio.Event()
         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._update_lock = asyncio.Lock()
         self._session_lock = asyncio.Lock()
 
@@ -465,7 +473,7 @@ class RealtimeSession(llm.RealtimeSession):
             finally:
                 await self._close_active_session()
 
-    async def _send_task(self, session: genai.LiveSession):
+    async def _send_task(self, session: AsyncSession):
         try:
             async for msg in self._msg_ch:
                 async with self._session_lock:
@@ -485,7 +493,7 @@ class RealtimeSession(llm.RealtimeSession):
         finally:
             logger.debug("send task finished.")
 
-    async def _recv_task(self, session: genai.LiveSession):
+    async def _recv_task(self, session: AsyncSession):
         try:
             while True:
                 async with self._session_lock:
@@ -501,6 +509,15 @@ class RealtimeSession(llm.RealtimeSession):
                     ):
                         self._start_new_generation()
 
+                    if response.session_resumption_update:
+                        if (
+                            response.session_resumption_update.resumable
+                            and response.session_resumption_update.new_handle
+                        ):
+                            self._session_resumption_handle = (
+                                response.session_resumption_update.new_handle
+                            )
+
                     if response.server_content:
                         self._handle_server_content(response.server_content)
                     if response.tool_call:
@@ -548,11 +565,13 @@ class RealtimeSession(llm.RealtimeSession):
             speech_config=SpeechConfig(
                 voice_config=VoiceConfig(
                     prebuilt_voice_config=PrebuiltVoiceConfig(voice_name=self._opts.voice)
-                )
+                ),
+                language_code=self._opts.language if is_given(self._opts.language) else None,
             ),
             tools=[Tool(function_declarations=self._gemini_declarations)],
             input_audio_transcription=self._opts.input_audio_transcription,
             output_audio_transcription=self._opts.output_audio_transcription,
+            session_resumption=SessionResumptionConfig(handle=self._session_resumption_handle),
         )
 
     def _start_new_generation(self):

livekit/plugins/google/tts.py

@@ -105,7 +105,7 @@ class TTS(tts.TTS):
         self._opts = _TTSOptions(
             voice=voice_params,
             audio_config=texttospeech.AudioConfig(
-                audio_encoding=texttospeech.AudioEncoding.OGG_OPUS,
+                audio_encoding=texttospeech.AudioEncoding.PCM,
                 sample_rate_hertz=sample_rate,
                 pitch=pitch,
                 effects_profile_id=effects_profile_id,
@@ -132,11 +132,11 @@ class TTS(tts.TTS):
         """  # noqa: E501
         params = {}
         if is_given(language):
-            params["language"] = language
+            params["language_code"] = str(language)
         if is_given(gender):
-            params["gender"] = gender
+            params["ssml_gender"] = _gender_from_str(str(gender))
         if is_given(voice_name):
-            params["voice_name"] = voice_name
+            params["name"] = voice_name
 
         if params:
             self._opts.voice = texttospeech.VoiceSelectionParams(**params)

livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.0.17"
+__version__ = "1.0.18"

{livekit_plugins_google-1.0.17.dist-info → livekit_plugins_google-1.0.18.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.0.17
+Version: 1.0.18
 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
-Requires-Dist: google-genai>=1.11.0
-Requires-Dist: livekit-agents>=1.0.17
+Requires-Dist: google-genai>=1.12.1
+Requires-Dist: livekit-agents>=1.0.18
 Description-Content-Type: text/markdown
 
 # LiveKit Plugins Google

{livekit_plugins_google-1.0.17.dist-info → livekit_plugins_google-1.0.18.dist-info}/RECORD

@@ -4,13 +4,13 @@ livekit/plugins/google/log.py,sha256=GI3YWN5YzrafnUccljzPRS_ZALkMNk1i21IRnTl2vNA
 livekit/plugins/google/models.py,sha256=SGjAumdDK97NNLwMFcqZdKR68f1NoGB2Rk1UP2-imG0,1457
 livekit/plugins/google/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 livekit/plugins/google/stt.py,sha256=AG_lh2fuuduJi0jFbA_QKFXLJ6NUdF1W_FfkLUJML_Q,22413
-livekit/plugins/google/tts.py,sha256=xhINokqY8UutXn85N-cbzq68eptbM6TTtIXmLktE_RM,9004
+livekit/plugins/google/tts.py,sha256=fmQwW9a1kPsEsrTvIo8fqw479RxWEx0SIc3oTVaj41U,9031
 livekit/plugins/google/utils.py,sha256=TjjTwMbdJdxr3bZjUXxs-J_fipTTM00goW2-d9KWX6w,9582
-livekit/plugins/google/version.py,sha256=GOfJB-DKZur-i3hrjFbzgpC2NHE96dnWhGLziW1e0_E,601
+livekit/plugins/google/version.py,sha256=cnPu9FVKZV9tFmmz7lEvftrO3B_nWJVFghi3j6UcJLs,601
 livekit/plugins/google/beta/__init__.py,sha256=AxRYc7NGG62Tv1MmcZVCDHNvlhbC86hM-_yP01Qb28k,47
 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=2_nPBvPttVudoQswhf19ieJ6wxvHquGJgALJ09afQms,29873
-livekit_plugins_google-1.0.17.dist-info/METADATA,sha256=cKeNSFwiM2A-MJeNA6zNeX7ioqbvkEZO3aFfR8Run2c,3492
-livekit_plugins_google-1.0.17.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_google-1.0.17.dist-info/RECORD,,
+livekit/plugins/google/beta/realtime/realtime_api.py,sha256=sXp2oHnTlHrAp5wFmcXj0bRtQKixBYedfbufcbjVHxk,30897
+livekit_plugins_google-1.0.18.dist-info/METADATA,sha256=Vqt0FoqibcKzX_jFXlyFkn-mT7iPC16JlH61VS0fbuw,3492
+livekit_plugins_google-1.0.18.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_google-1.0.18.dist-info/RECORD,,