livekit-plugins-google 1.2.8__py3-none-any.whl → 1.2.11__py3-none-any.whl

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

@@ -202,6 +202,7 @@ class RealtimeModel(llm.RealtimeModel):
                 user_transcription=input_audio_transcription is not None,
                 auto_tool_reply_generation=True,
                 audio_output=types.Modality.AUDIO in modalities,
+                manual_function_calls=False,
             )
         )
 
@@ -304,6 +305,10 @@ class RealtimeModel(llm.RealtimeModel):
     async def aclose(self) -> None:
         pass
 
+    @property
+    def model(self) -> str:
+        return self._opts.model
+
 
 class RealtimeSession(llm.RealtimeSession):
     def __init__(self, realtime_model: RealtimeModel) -> None:
@@ -775,7 +780,7 @@ class RealtimeSession(llm.RealtimeSession):
                 ),
                 language_code=self._opts.language if is_given(self._opts.language) else None,
             ),
-            tools=tools_config,  # type: ignore
+            tools=tools_config,
             input_audio_transcription=self._opts.input_audio_transcription,
             output_audio_transcription=self._opts.output_audio_transcription,
             session_resumption=types.SessionResumptionConfig(
@@ -829,6 +834,7 @@ class RealtimeSession(llm.RealtimeSession):
             message_stream=self._current_generation.message_ch,
             function_stream=self._current_generation.function_ch,
             user_initiated=False,
+            response_id=self._current_generation.response_id,
         )
 
         if self._pending_generation_fut and not self._pending_generation_fut.done():
@@ -969,7 +975,7 @@ class RealtimeSession(llm.RealtimeSession):
             gen.function_ch.send_nowait(
                 llm.FunctionCall(
                     call_id=fnc_call.id or utils.shortuuid("fnc-call-"),
-                    name=fnc_call.name,  # type: ignore
+                    name=fnc_call.name,
                     arguments=arguments,
                 )
             )
@@ -1018,7 +1024,8 @@ class RealtimeSession(llm.RealtimeSession):
             return token_details_map
 
         metrics = RealtimeModelMetrics(
-            label=self._realtime_model._label,
+            label=self._realtime_model.label,
+            model=self._realtime_model.model,
             request_id=current_gen.response_id,
             timestamp=current_gen._created_timestamp,
             duration=duration,

livekit/plugins/google/llm.py

@@ -423,7 +423,7 @@ class LLMStream(llm.LLMStream):
                     tool_calls=[
                         llm.FunctionToolCall(
                             arguments=json.dumps(part.function_call.args),
-                            name=part.function_call.name,  # type: ignore
+                            name=part.function_call.name,
                             call_id=part.function_call.id or utils.shortuuid("function_call_"),
                         )
                     ],

livekit/plugins/google/stt.py

@@ -70,6 +70,7 @@ class STTOptions:
     spoken_punctuation: bool
     enable_word_time_offsets: bool
     enable_word_confidence: bool
+    enable_voice_activity_events: bool
     model: SpeechModels | str
     sample_rate: int
     min_confidence_threshold: float
@@ -103,6 +104,7 @@ class STT(stt.STT):
         spoken_punctuation: bool = False,
         enable_word_time_offsets: bool = True,
         enable_word_confidence: bool = False,
+        enable_voice_activity_events: bool = False,
         model: SpeechModels | str = "latest_long",
         location: str = "global",
         sample_rate: int = 16000,
@@ -127,6 +129,7 @@ class STT(stt.STT):
             spoken_punctuation(bool): whether to use spoken punctuation (default: False)
             enable_word_time_offsets(bool): whether to enable word time offsets (default: True)
             enable_word_confidence(bool): whether to enable word confidence (default: False)
+            enable_voice_activity_events(bool): whether to enable voice activity events (default: False)
             model(SpeechModels): the model to use for recognition default: "latest_long"
             location(str): the location to use for recognition default: "global"
             sample_rate(int): the sample rate of the audio default: 16000
@@ -168,6 +171,7 @@ class STT(stt.STT):
             spoken_punctuation=spoken_punctuation,
             enable_word_time_offsets=enable_word_time_offsets,
             enable_word_confidence=enable_word_confidence,
+            enable_voice_activity_events=enable_voice_activity_events,
             model=model,
             sample_rate=sample_rate,
             min_confidence_threshold=min_confidence_threshold,
@@ -507,6 +511,7 @@ class SpeechStream(stt.SpeechStream):
                         ),
                         streaming_features=cloud_speech.StreamingRecognitionFeatures(
                             interim_results=self._config.interim_results,
+                            enable_voice_activity_events=self._config.enable_voice_activity_events,
                         ),
                     )
 

livekit/plugins/google/tts.py

@@ -52,6 +52,7 @@ class _TTSOptions:
     volume_gain_db: float
     custom_pronunciations: CustomPronunciations | None
     enable_ssml: bool
+    use_markup: bool
 
 
 class TTS(tts.TTS):
@@ -75,6 +76,7 @@ class TTS(tts.TTS):
         custom_pronunciations: NotGivenOr[CustomPronunciations] = NOT_GIVEN,
         use_streaming: bool = True,
         enable_ssml: bool = False,
+        use_markup: bool = False,
     ) -> None:
         """
         Create a new instance of Google TTS.
@@ -100,6 +102,7 @@ class TTS(tts.TTS):
             custom_pronunciations (CustomPronunciations, optional): Custom pronunciations for the TTS. Default is None.
             use_streaming (bool, optional): Whether to use streaming synthesis. Default is True.
             enable_ssml (bool, optional): Whether to enable SSML support. Default is False.
+            use_markup (bool, optional): Whether to enable markup input for HD voices. Default is False.
         """  # noqa: E501
         super().__init__(
             capabilities=tts.TTSCapabilities(streaming=use_streaming),
@@ -107,8 +110,11 @@ class TTS(tts.TTS):
             num_channels=1,
         )
 
-        if enable_ssml and use_streaming:
-            raise ValueError("SSML support is not available for streaming synthesis")
+        if enable_ssml:
+            if use_streaming:
+                raise ValueError("SSML support is not available for streaming synthesis")
+            if use_markup:
+                raise ValueError("SSML support is not available for markup input")
 
         self._client: texttospeech.TextToSpeechAsyncClient | None = None
         self._credentials_info = credentials_info
@@ -145,6 +151,7 @@ class TTS(tts.TTS):
             volume_gain_db=volume_gain_db,
             custom_pronunciations=pronunciations,
             enable_ssml=enable_ssml,
+            use_markup=use_markup,
         )
         self._streams = weakref.WeakSet[SynthesizeStream]()
 
@@ -238,19 +245,21 @@ class ChunkedStream(tts.ChunkedStream):
 
     async def _run(self, output_emitter: tts.AudioEmitter) -> None:
         try:
-            input = (
-                texttospeech.SynthesisInput(
-                    ssml=self._build_ssml(),
-                    custom_pronunciations=self._opts.custom_pronunciations,
+            if self._opts.use_markup:
+                tts_input = texttospeech.SynthesisInput(
+                    markup=self._input_text, custom_pronunciations=self._opts.custom_pronunciations
                 )
-                if self._opts.enable_ssml
-                else texttospeech.SynthesisInput(
-                    text=self._input_text,
-                    custom_pronunciations=self._opts.custom_pronunciations,
+            elif self._opts.enable_ssml:
+                tts_input = texttospeech.SynthesisInput(
+                    ssml=self._build_ssml(), custom_pronunciations=self._opts.custom_pronunciations
                 )
-            )
+            else:
+                tts_input = texttospeech.SynthesisInput(
+                    text=self._input_text, custom_pronunciations=self._opts.custom_pronunciations
+                )
+
             response: SynthesizeSpeechResponse = await self._tts._ensure_client().synthesize_speech(
-                input=input,
+                input=tts_input,
                 voice=self._opts.voice,
                 audio_config=texttospeech.AudioConfig(
                     audio_encoding=self._opts.encoding,
@@ -355,8 +364,12 @@ class SynthesizeStream(tts.SynthesizeStream):
 
                 async for input in input_stream:
                     self._mark_started()
-                    yield texttospeech.StreamingSynthesizeRequest(
-                        input=texttospeech.StreamingSynthesisInput(text=input.token)
+                    yield (
+                        texttospeech.StreamingSynthesizeRequest(
+                            input=texttospeech.StreamingSynthesisInput(markup=input.token)
+                            if self._opts.use_markup
+                            else texttospeech.StreamingSynthesisInput(text=input.token)
+                        )
                     )
 
             except Exception:

livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.2.8"
+__version__ = "1.2.11"

{livekit_plugins_google-1.2.8.dist-info → livekit_plugins_google-1.2.11.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.2.8
+Version: 1.2.11
 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>=1.2.8
+Requires-Dist: livekit-agents>=1.2.11
 Description-Content-Type: text/markdown
 
 # Google AI plugin for LiveKit Agents

{livekit_plugins_google-1.2.8.dist-info → livekit_plugins_google-1.2.11.dist-info}/RECORD

@@ -1,18 +1,18 @@
 livekit/plugins/google/__init__.py,sha256=XIyZ-iFnRBpaLtOJgVwojlB-a8GjdDugVFcjBpMEww8,1412
-livekit/plugins/google/llm.py,sha256=cMlmLX1m3TsrLW0a-k2oj6WQSNWEjj3jv7ob8MUoXCI,18825
+livekit/plugins/google/llm.py,sha256=aeeGqhbEScbEs-GKp1T8rLocNqmvG4UBj6diekYe4FU,18809
 livekit/plugins/google/log.py,sha256=GI3YWN5YzrafnUccljzPRS_ZALkMNk1i21IRnTl2vNA,69
 livekit/plugins/google/models.py,sha256=poOvUBvgpqmmQV5EUQsq0RgNIRAq7nH-_IZIcIfPSBI,2801
 livekit/plugins/google/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-livekit/plugins/google/stt.py,sha256=gRhVRsfg3BPNkBJGG78QOxEia0mF1jBnI_Ckq1jxqIs,25938
+livekit/plugins/google/stt.py,sha256=RhbDkCbrGWb8R7feS5w766fwUNJEbg7hHlba7Oq9lBI,26305
 livekit/plugins/google/tools.py,sha256=tD5HVDHO5JfUF029Cx3axHMJec0Gxalkl7s1FDgxLzI,259
-livekit/plugins/google/tts.py,sha256=3TPHBKJJwIt-hSTAdbI4NUcQNerhV0eDuK_o2rprdqg,16606
+livekit/plugins/google/tts.py,sha256=LBLP3pEq1iCCgfidpRTtpeoDKYmXh8PKeJf1llAsybQ,17302
 livekit/plugins/google/utils.py,sha256=z0iCP6-hYix3JRm2RM5JOBEJCICehUe5N4FTl-JpXLc,9269
-livekit/plugins/google/version.py,sha256=XZ3forlpqz9F1SmliM6XQGe2MyHu60jOjDbNGs_mcRg,600
+livekit/plugins/google/version.py,sha256=xsWwuH5qgJrB3wPPfmZaiEH7zObN2yGboBmyrTHj-b8,601
 livekit/plugins/google/beta/__init__.py,sha256=RvAUdvEiRN-fe4JrgPcN0Jkw1kZR9wPerGMFVjS1Cc0,270
 livekit/plugins/google/beta/gemini_tts.py,sha256=esWjr0Xf95tl0_AB7MXiFZ_VCORWgcWjzvLvRa3t0FQ,8515
 livekit/plugins/google/beta/realtime/__init__.py,sha256=_fW2NMN22F-hnQ4xAJ_g5lPbR7CvM_xXzSWlUQY-E-U,188
 livekit/plugins/google/beta/realtime/api_proto.py,sha256=nb_QkVQDEH7h0SKA9vdS3JaL12a6t2Z1ja4SdnxE6a8,814
-livekit/plugins/google/beta/realtime/realtime_api.py,sha256=p0vEaxQhPLUbGjHo7Za2rbBrCjD_UqPk-thd9ybIiuk,47817
-livekit_plugins_google-1.2.8.dist-info/METADATA,sha256=2I3YmnnGQGcd8qxz8AGaxG_KflpdYu-3oUBVCudZfA8,1907
-livekit_plugins_google-1.2.8.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_google-1.2.8.dist-info/RECORD,,
+livekit/plugins/google/beta/realtime/realtime_api.py,sha256=RALLfKWb8c4K8ennINDLeVxKrP5JXvGa_nNGP0_ASlI,48012
+livekit_plugins_google-1.2.11.dist-info/METADATA,sha256=ulwiM6njmKCQG_8e1imKwV6oG0IHuXRahBSB0UI0OBM,1909
+livekit_plugins_google-1.2.11.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_google-1.2.11.dist-info/RECORD,,