livekit-plugins-google 1.1.4__tar.gz → 1.1.6__tar.gz

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-google
-Version: 1.1.4
+Version: 1.1.6
 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,>=2.27
-Requires-Dist: google-genai>=v1.21.1
-Requires-Dist: livekit-agents>=1.1.4
+Requires-Dist: google-genai>=v1.23.0
+Requires-Dist: livekit-agents>=1.1.6
 Description-Content-Type: text/markdown
 
 # Google AI plugin for LiveKit Agents

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/livekit/plugins/google/beta/realtime/api_proto.py

@@ -13,7 +13,7 @@ LiveAPIModels = Literal[
     "gemini-2.5-flash-exp-native-audio-thinking-dialog",
 ]
 
-Voice = Literal["Puck", "Charon", "Kore", "Fenrir", "Aoede", "Leda", "Oru", "Zephyr"]
+Voice = Literal["Puck", "Charon", "Kore", "Fenrir", "Aoede", "Leda", "Orus", "Zephyr"]
 
 
 ClientEvents = Union[

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/livekit/plugins/google/llm.py

@@ -62,6 +62,7 @@ class _LLMOptions:
     automatic_function_calling_config: NotGivenOr[types.AutomaticFunctionCallingConfigOrDict]
     gemini_tools: NotGivenOr[list[_LLMTool]]
     http_options: NotGivenOr[types.HttpOptions]
+    seed: NotGivenOr[int]
 
 
 class LLM(llm.LLM):
@@ -86,6 +87,7 @@ class LLM(llm.LLM):
         ] = NOT_GIVEN,
         gemini_tools: NotGivenOr[list[_LLMTool]] = NOT_GIVEN,
         http_options: NotGivenOr[types.HttpOptions] = NOT_GIVEN,
+        seed: NotGivenOr[int] = NOT_GIVEN,
     ) -> None:
         """
         Create a new instance of Google GenAI LLM.
@@ -176,6 +178,7 @@ class LLM(llm.LLM):
             automatic_function_calling_config=automatic_function_calling_config,
             gemini_tools=gemini_tools,
             http_options=http_options,
+            seed=seed,
         )
         self._client = Client(
             api_key=gemini_api_key,
@@ -262,6 +265,8 @@ class LLM(llm.LLM):
             extra["presence_penalty"] = self._opts.presence_penalty
         if is_given(self._opts.frequency_penalty):
             extra["frequency_penalty"] = self._opts.frequency_penalty
+        if is_given(self._opts.seed):
+            extra["seed"] = self._opts.seed
 
         # Add thinking config if thinking_budget is provided
         if is_given(self._opts.thinking_config):

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/livekit/plugins/google/tts.py

@@ -22,7 +22,11 @@ from dataclasses import dataclass, replace
 from google.api_core.client_options import ClientOptions
 from google.api_core.exceptions import DeadlineExceeded, GoogleAPICallError
 from google.cloud import texttospeech
-from google.cloud.texttospeech_v1.types import SsmlVoiceGender, SynthesizeSpeechResponse
+from google.cloud.texttospeech_v1.types import (
+    CustomPronunciations,
+    SsmlVoiceGender,
+    SynthesizeSpeechResponse,
+)
 from livekit.agents import APIConnectOptions, APIStatusError, APITimeoutError, tokenize, tts, utils
 from livekit.agents.types import DEFAULT_API_CONNECT_OPTIONS, NOT_GIVEN, NotGivenOr
 from livekit.agents.utils import is_given
@@ -47,6 +51,7 @@ class _TTSOptions:
     speaking_rate: float
     tokenizer: tokenize.SentenceTokenizer
     volume_gain_db: float
+    custom_pronunciations: CustomPronunciations | None
     enable_ssml: bool
 
 
@@ -67,6 +72,7 @@ class TTS(tts.TTS):
         credentials_info: NotGivenOr[dict] = NOT_GIVEN,
         credentials_file: NotGivenOr[str] = NOT_GIVEN,
         tokenizer: NotGivenOr[tokenize.SentenceTokenizer] = NOT_GIVEN,
+        custom_pronunciations: NotGivenOr[CustomPronunciations] = NOT_GIVEN,
         use_streaming: bool = True,
         enable_ssml: bool = False,
     ) -> None:
@@ -90,6 +96,7 @@ class TTS(tts.TTS):
             credentials_info (dict, optional): Dictionary containing Google Cloud credentials. Default is None.
             credentials_file (str, optional): Path to the Google Cloud credentials JSON file. Default is None.
             tokenizer (tokenize.SentenceTokenizer, optional): Tokenizer for the TTS. Default is a basic sentence tokenizer.
+            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.
         """  # noqa: E501
@@ -119,6 +126,8 @@ class TTS(tts.TTS):
         if not is_given(tokenizer):
             tokenizer = tokenize.basic.SentenceTokenizer(min_sentence_len=BUFFERED_WORDS_COUNT)
 
+        pronunciations = None if not is_given(custom_pronunciations) else custom_pronunciations
+
         self._opts = _TTSOptions(
             voice=voice_params,
             encoding=audio_encoding,
@@ -128,6 +137,7 @@ class TTS(tts.TTS):
             speaking_rate=speaking_rate,
             tokenizer=tokenizer,
             volume_gain_db=volume_gain_db,
+            custom_pronunciations=pronunciations,
             enable_ssml=enable_ssml,
         )
         self._streams = weakref.WeakSet[SynthesizeStream]()
@@ -223,9 +233,15 @@ class ChunkedStream(tts.ChunkedStream):
     async def _run(self, output_emitter: tts.AudioEmitter) -> None:
         try:
             input = (
-                texttospeech.SynthesisInput(ssml=self._build_ssml())
+                texttospeech.SynthesisInput(
+                    ssml=self._build_ssml(),
+                    custom_pronunciations=self._opts.custom_pronunciations,
+                )
                 if self._opts.enable_ssml
-                else texttospeech.SynthesisInput(text=self._input_text)
+                else texttospeech.SynthesisInput(
+                    text=self._input_text,
+                    custom_pronunciations=self._opts.custom_pronunciations,
+                )
             )
             response: SynthesizeSpeechResponse = await self._tts._ensure_client().synthesize_speech(
                 input=input,
@@ -287,6 +303,7 @@ class SynthesizeStream(tts.SynthesizeStream):
                 sample_rate_hertz=self._opts.sample_rate,
                 speaking_rate=self._opts.speaking_rate,
             ),
+            custom_pronunciations=self._opts.custom_pronunciations,
         )
 
         async def _tokenize_input() -> None:

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.1.4"
+__version__ = "1.1.6"

{livekit_plugins_google-1.1.4 → livekit_plugins_google-1.1.6}/pyproject.toml

@@ -26,8 +26,8 @@ dependencies = [
     "google-auth >= 2, < 3",
     "google-cloud-speech >= 2, < 3",
     "google-cloud-texttospeech >= 2.27, < 3",
-    "google-genai >= v1.21.1",
-    "livekit-agents>=1.1.4",
+    "google-genai >= v1.23.0",
+    "livekit-agents>=1.1.6",
 ]
 
 [project.urls]