livekit-plugins-google 0.6.2__py3-none-any.whl → 0.7.0__py3-none-any.whl

livekit/plugins/google/stt.py

@@ -16,13 +16,14 @@ from __future__ import annotations
 
 import asyncio
 import dataclasses
-import os
 from dataclasses import dataclass
 from typing import AsyncIterable, List, Union
 
 from livekit import agents, rtc
 from livekit.agents import stt, utils
 
+from google.auth import default as gauth_default
+from google.auth.exceptions import DefaultCredentialsError
 from google.cloud.speech_v2 import SpeechAsyncClient
 from google.cloud.speech_v2.types import cloud_speech
 
@@ -58,8 +59,11 @@ class STT(stt.STT):
         credentials_file: str | None = None,
     ):
         """
-        if no credentials is provided, it will use the credentials on the environment
-        GOOGLE_APPLICATION_CREDENTIALS (default behavior of Google SpeechAsyncClient)
+        Create a new instance of Google STT.
+
+        Credentials must be provided, either by using the ``credentials_info`` dict, or reading
+        from the file specified in ``credentials_file`` or via Application Default Credentials as
+        described in https://cloud.google.com/docs/authentication/application-default-credentials
         """
         super().__init__(
             capabilities=stt.STTCapabilities(streaming=True, interim_results=True)
@@ -70,10 +74,13 @@ class STT(stt.STT):
         self._credentials_file = credentials_file
 
         if credentials_file is None and credentials_info is None:
-            creds = os.environ.get("GOOGLE_APPLICATION_CREDENTIALS")
-            if not creds:
+            try:
+                gauth_default()
+            except DefaultCredentialsError:
                 raise ValueError(
-                    "GOOGLE_APPLICATION_CREDENTIALS must be set if no credentials is provided"
+                    "Application default credentials must be available "
+                    "when using Google STT without explicitly passing "
+                    "credentials through credentials_info or credentials_file."
                 )
 
         if isinstance(languages, str):
@@ -109,7 +116,12 @@ class STT(stt.STT):
         # recognizers may improve latency https://cloud.google.com/speech-to-text/v2/docs/recognizers#understand_recognizers
 
         # TODO(theomonnom): find a better way to access the project_id
-        project_id = self._ensure_client().transport._credentials.project_id  # type: ignore
+        try:
+            project_id = self._ensure_client().transport._credentials.project_id  # type: ignore
+        except AttributeError:
+            from google.auth import default as ga_default
+
+            _, project_id = ga_default()
         return f"projects/{project_id}/locations/global/recognizers/_"
 
     def _sanitize_options(self, *, language: str | None = None) -> STTOptions:
@@ -278,22 +290,22 @@ class SpeechStream(stt.SpeechStream):
                 == cloud_speech.StreamingRecognizeResponse.SpeechEventType.SPEECH_EVENT_TYPE_UNSPECIFIED
             ):
                 result = resp.results[0]
+                speech_data = _streaming_recognize_response_to_speech_data(resp)
+                if speech_data is None:
+                    continue
+
                 if not result.is_final:
                     self._event_ch.send_nowait(
                         stt.SpeechEvent(
                             type=stt.SpeechEventType.INTERIM_TRANSCRIPT,
-                            alternatives=[
-                                _streaming_recognize_response_to_speech_data(resp)
-                            ],
+                            alternatives=[speech_data],
                         )
                     )
                 else:
                     self._event_ch.send_nowait(
                         stt.SpeechEvent(
                             type=stt.SpeechEventType.FINAL_TRANSCRIPT,
-                            alternatives=[
-                                _streaming_recognize_response_to_speech_data(resp)
-                            ],
+                            alternatives=[speech_data],
                         )
                     )
 
@@ -337,16 +349,21 @@ def _recognize_response_to_speech_event(
 
 def _streaming_recognize_response_to_speech_data(
     resp: cloud_speech.StreamingRecognizeResponse,
-) -> stt.SpeechData:
+) -> stt.SpeechData | None:
     text = ""
     confidence = 0.0
     for result in resp.results:
+        if len(result.alternatives) == 0:
+            continue
         text += result.alternatives[0].transcript
         confidence += result.alternatives[0].confidence
 
     confidence /= len(resp.results)
     lg = resp.results[0].language_code
 
+    if text == "":
+        return None
+
     data = stt.SpeechData(
         language=lg, start_time=0, end_time=0, confidence=confidence, text=text
     )

livekit/plugins/google/tts.py

@@ -51,9 +51,13 @@ class TTS(tts.TTS):
         credentials_file: str | None = None,
     ) -> None:
         """
-        if no credentials is provided, it will use the credentials on the environment
-        GOOGLE_APPLICATION_CREDENTIALS (default behavior of Google TextToSpeechAsyncClient)
+        Create a new instance of Google TTS.
+
+        Credentials must be provided, either by using the ``credentials_info`` dict, or reading
+        from the file specified in ``credentials_file`` or the ``GOOGLE_APPLICATION_CREDENTIALS``
+        environmental variable.
         """
+
         super().__init__(
             capabilities=tts.TTSCapabilities(
                 streaming=False,
@@ -144,6 +148,7 @@ class ChunkedStream(tts.ChunkedStream):
                     )
                 )
         else:
+            data = data[44:]  # skip WAV header
             self._event_ch.send_nowait(
                 tts.SynthesizedAudio(
                     request_id=request_id,

livekit/plugins/google/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "0.6.2"
+__version__ = "0.7.0"

{livekit_plugins_google-0.6.2.dist-info → livekit_plugins_google-0.7.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: livekit-plugins-google
-Version: 0.6.2
+Version: 0.7.0
 Summary: Agent Framework plugin for services from Google Cloud
 Home-page: https://github.com/livekit/agents
 License: Apache-2.0
@@ -19,6 +19,7 @@ Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3 :: Only
 Requires-Python: >=3.9.0
 Description-Content-Type: text/markdown
+Requires-Dist: google-auth <3,>=2
 Requires-Dist: google-cloud-speech <3,>=2
 Requires-Dist: google-cloud-texttospeech <3,>=2
 Requires-Dist: livekit-agents >=0.8.0.dev0
@@ -35,4 +36,4 @@ pip install livekit-plugins-google
 
 ## Pre-requisites
 
-For credentials, you'll need a Google Cloud account and obtain the correct credentials. Credentials can be passed directly or set as [GOOGLE_APPLICATION_CREDENTIALS](https://cloud.google.com/docs/authentication/application-default-credentials) environment variable.
+For credentials, you'll need a Google Cloud account and obtain the correct credentials. Credentials can be passed directly or via Application Default Credentials as specified in [How Application Default Credentials works](https://cloud.google.com/docs/authentication/application-default-credentials).

livekit_plugins_google-0.7.0.dist-info/RECORD

@@ -0,0 +1,11 @@
+livekit/plugins/google/__init__.py,sha256=CYbSmm5fEw71F_r_4pEApGaWQ_r15Y3ZEocH88a4yc8,948
+livekit/plugins/google/log.py,sha256=GI3YWN5YzrafnUccljzPRS_ZALkMNk1i21IRnTl2vNA,69
+livekit/plugins/google/models.py,sha256=n8pgTJ7xyJpPCZJ_y0GzaQq6LqYknL6K6trpi07-AxM,1307
+livekit/plugins/google/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+livekit/plugins/google/stt.py,sha256=XXDOISg-8U1MzVu543xLEB3-mr_NFKJp9qo1-ya2-Hc,13569
+livekit/plugins/google/tts.py,sha256=LQttOY3rI8TQ7w3FT-nBv-PDg5oXwITvFeBZtjwrwJE,5692
+livekit/plugins/google/version.py,sha256=G63knoeV7ai0fH-1DCHqI3a7eSI4LlHqjV64n4GbCGg,600
+livekit_plugins_google-0.7.0.dist-info/METADATA,sha256=cHccq2kH8vnXd7qNkADTDbmH1dOd7haUonMGNvdMbmo,1653
+livekit_plugins_google-0.7.0.dist-info/WHEEL,sha256=cVxcB9AmuTcXqmwrtPhNK88dr7IR_b6qagTj0UvIEbY,91
+livekit_plugins_google-0.7.0.dist-info/top_level.txt,sha256=OoDok3xUmXbZRvOrfvvXB-Juu4DX79dlq188E19YHoo,8
+livekit_plugins_google-0.7.0.dist-info/RECORD,,

{livekit_plugins_google-0.6.2.dist-info → livekit_plugins_google-0.7.0.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (72.1.0)
+Generator: setuptools (74.1.2)
 Root-Is-Purelib: true
 Tag: py3-none-any
 

livekit_plugins_google-0.6.2.dist-info/RECORD

@@ -1,11 +0,0 @@
-livekit/plugins/google/__init__.py,sha256=CYbSmm5fEw71F_r_4pEApGaWQ_r15Y3ZEocH88a4yc8,948
-livekit/plugins/google/log.py,sha256=GI3YWN5YzrafnUccljzPRS_ZALkMNk1i21IRnTl2vNA,69
-livekit/plugins/google/models.py,sha256=n8pgTJ7xyJpPCZJ_y0GzaQq6LqYknL6K6trpi07-AxM,1307
-livekit/plugins/google/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-livekit/plugins/google/stt.py,sha256=Jt3_uc6F9rBZootKyPYslgCgusQB_k7NQ1Cvj9CwppI,12970
-livekit/plugins/google/tts.py,sha256=jvbw-T-JlVxcYWiF-tCox35LZuCN3DBKwDp3zN8JCTw,5549
-livekit/plugins/google/version.py,sha256=hj5La4IQz5ccAWt5oJAkV9TnNFuujYmxSTjcRby-kNQ,600
-livekit_plugins_google-0.6.2.dist-info/METADATA,sha256=fvHmaZHNW-dAFBr8-G5Mrm1sHs66AY6Ur8rXvD2Q-rg,1584
-livekit_plugins_google-0.6.2.dist-info/WHEEL,sha256=R0nc6qTxuoLk7ShA2_Y-UWkN8ZdfDBG2B6Eqpz2WXbs,91
-livekit_plugins_google-0.6.2.dist-info/top_level.txt,sha256=OoDok3xUmXbZRvOrfvvXB-Juu4DX79dlq188E19YHoo,8
-livekit_plugins_google-0.6.2.dist-info/RECORD,,