livekit-plugins-azure 1.0.0rc9__py3-none-any.whl → 1.0.2__py3-none-any.whl

livekit/plugins/azure/stt.py

@@ -21,7 +21,13 @@ from dataclasses import dataclass
 
 import azure.cognitiveservices.speech as speechsdk  # type: ignore
 from livekit import rtc
-from livekit.agents import DEFAULT_API_CONNECT_OPTIONS, APIConnectOptions, stt, utils
+from livekit.agents import (
+    DEFAULT_API_CONNECT_OPTIONS,
+    APIConnectionError,
+    APIConnectOptions,
+    stt,
+    utils,
+)
 from livekit.agents.types import (
     NOT_GIVEN,
     NotGivenOr,
@@ -65,12 +71,15 @@ class STT(stt.STT):
         # Azure handles multiple languages and can auto-detect the language used. It requires the candidate set to be set.  # noqa: E501
         language: NotGivenOr[str | list[str] | None] = NOT_GIVEN,
         profanity: NotGivenOr[speechsdk.enums.ProfanityOption] = NOT_GIVEN,
+        speech_endpoint: NotGivenOr[str] = NOT_GIVEN,
     ):
         """
         Create a new instance of Azure STT.
 
         Either ``speech_host`` or ``speech_key`` and ``speech_region`` or
-        ``speech_auth_token`` and ``speech_region`` must be set using arguments.
+        ``speech_auth_token`` and ``speech_region`` or
+        ``speech_key`` and ``speech_endpoint``
+        must be set using arguments.
          Alternatively,  set the ``AZURE_SPEECH_HOST``, ``AZURE_SPEECH_KEY``
         and ``AZURE_SPEECH_REGION`` environmental variables, respectively.
         ``speech_auth_token`` must be set using the arguments as it's an ephemeral token.
@@ -96,9 +105,10 @@ class STT(stt.STT):
             is_given(speech_host)
             or (is_given(speech_key) and is_given(speech_region))
             or (is_given(speech_auth_token) and is_given(speech_region))
+            or (is_given(speech_key) and is_given(speech_endpoint))
         ):
             raise ValueError(
-                "AZURE_SPEECH_HOST or AZURE_SPEECH_KEY and AZURE_SPEECH_REGION or speech_auth_token and AZURE_SPEECH_REGION must be set"  # noqa: E501
+                "AZURE_SPEECH_HOST or AZURE_SPEECH_KEY and AZURE_SPEECH_REGION or speech_auth_token and AZURE_SPEECH_REGION or AZURE_SPEECH_KEY and speech_endpoint must be set"  # noqa: E501
             )
 
         self._config = STTOptions(
@@ -113,6 +123,7 @@ class STT(stt.STT):
             segmentation_max_time_ms=segmentation_max_time_ms,
             segmentation_strategy=segmentation_strategy,
             profanity=profanity,
+            speech_endpoint=speech_endpoint,
         )
         self._streams = weakref.WeakSet[SpeechStream]()
 
@@ -193,15 +204,20 @@ class SpeechStream(stt.SpeechStream):
 
                 process_input_task = asyncio.create_task(process_input())
                 wait_reconnect_task = asyncio.create_task(self._reconnect_event.wait())
+                wait_stopped_task = asyncio.create_task(self._session_stopped_event.wait())
 
                 try:
                     done, _ = await asyncio.wait(
-                        [process_input_task, wait_reconnect_task],
+                        [process_input_task, wait_reconnect_task, wait_stopped_task],
                         return_when=asyncio.FIRST_COMPLETED,
                     )
                     for task in done:
-                        if task != wait_reconnect_task:
+                        if task not in [wait_reconnect_task, wait_stopped_task]:
                             task.result()
+
+                    if wait_stopped_task in done:
+                        raise APIConnectionError("SpeechRecognition session stopped")
+
                     if wait_reconnect_task not in done:
                         break
                     self._reconnect_event.clear()

livekit/plugins/azure/tts.py

@@ -172,6 +172,7 @@ class TTS(tts.TTS):
         on_synthesizing_event: NotGivenOr[Callable] = NOT_GIVEN,
         on_viseme_event: NotGivenOr[Callable] = NOT_GIVEN,
         on_word_boundary_event: NotGivenOr[Callable] = NOT_GIVEN,
+        speech_endpoint: NotGivenOr[str] = NOT_GIVEN,
     ) -> None:
         """
         Create a new instance of Azure TTS.
@@ -209,9 +210,10 @@ class TTS(tts.TTS):
             is_given(speech_host)
             or (is_given(speech_key) and is_given(speech_region))
             or (is_given(speech_auth_token) and is_given(speech_region))
+            or (is_given(speech_key) and is_given(speech_endpoint))
         ):
             raise ValueError(
-                "AZURE_SPEECH_HOST or AZURE_SPEECH_KEY and AZURE_SPEECH_REGION or speech_auth_token and AZURE_SPEECH_REGION must be set"  # noqa: E501
+                "AZURE_SPEECH_HOST or AZURE_SPEECH_KEY and AZURE_SPEECH_REGION or speech_auth_token and AZURE_SPEECH_REGION or AZURE_SPEECH_KEY and speech_endpoint must be set"  # noqa: E501
             )
 
         if is_given(prosody):
@@ -238,6 +240,7 @@ class TTS(tts.TTS):
             on_synthesizing_event=on_synthesizing_event,
             on_viseme_event=on_viseme_event,
             on_word_boundary_event=on_word_boundary_event,
+            speech_endpoint=speech_endpoint,
         )
 
     def update_options(

livekit/plugins/azure/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = '1.0.0.rc9'
+__version__ = '1.0.2'

{livekit_plugins_azure-1.0.0rc9.dist-info → livekit_plugins_azure-1.0.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-azure
-Version: 1.0.0rc9
+Version: 1.0.2
 Summary: Agent Framework plugin for services from Azure
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
@@ -19,7 +19,7 @@ Classifier: Topic :: Multimedia :: Video
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.9.0
 Requires-Dist: azure-cognitiveservices-speech>=1.41.0
-Requires-Dist: livekit-agents>=1.0.0.rc9
+Requires-Dist: livekit-agents>=1.0.2
 Description-Content-Type: text/markdown
 
 # LiveKit Plugins Azure

livekit_plugins_azure-1.0.2.dist-info/RECORD

@@ -0,0 +1,9 @@
+livekit/plugins/azure/__init__.py,sha256=8YPW1EAltd4oip3WItsMy17EMekroUzTPLFRIwLP2Wc,1082
+livekit/plugins/azure/log.py,sha256=MeD0unQJ72aDc9K8zUi9LgUBls6h2WUALryOjAumrKs,68
+livekit/plugins/azure/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+livekit/plugins/azure/stt.py,sha256=vEB5w03F4dyu_BxQ849JMRgGuzTMNmtLDlUvWi44fvQ,14150
+livekit/plugins/azure/tts.py,sha256=e4BccMEnx7Qukij5t1H47w6XzFLeZIPlUVhjZ5yJAFg,19280
+livekit/plugins/azure/version.py,sha256=RSUx0PwRR34jaRRP_sG9HJWmSKxdazVULy4BLENZ3S8,600
+livekit_plugins_azure-1.0.2.dist-info/METADATA,sha256=qGEOeAh85HNRVZvosLOlUHwwBip3xP-aTF4P6FiAdv0,1394
+livekit_plugins_azure-1.0.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_azure-1.0.2.dist-info/RECORD,,

livekit_plugins_azure-1.0.0rc9.dist-info/RECORD

@@ -1,9 +0,0 @@
-livekit/plugins/azure/__init__.py,sha256=8YPW1EAltd4oip3WItsMy17EMekroUzTPLFRIwLP2Wc,1082
-livekit/plugins/azure/log.py,sha256=MeD0unQJ72aDc9K8zUi9LgUBls6h2WUALryOjAumrKs,68
-livekit/plugins/azure/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-livekit/plugins/azure/stt.py,sha256=TdjTMwSNlVBnrY6IB8ndS7OLuVOJtV1o3O7O4Vlnmsw,13566
-livekit/plugins/azure/tts.py,sha256=A2x7f1iS7dTZij1K7FoVILynfU1YGQiwG5venknrfXo,19073
-livekit/plugins/azure/version.py,sha256=_0m64Sw2cs3YI-DUoRXhekToRoCXFqqShFV-7Dvs3jo,604
-livekit_plugins_azure-1.0.0rc9.dist-info/METADATA,sha256=Ihk8M4HZfDbUOZgNjeTX7UHcZMRJWkDjsVaMQEimbs8,1401
-livekit_plugins_azure-1.0.0rc9.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_azure-1.0.0rc9.dist-info/RECORD,,