livekit-plugins-speechmatics 1.0.23__py3-none-any.whl → 1.1.0__py3-none-any.whl

livekit/plugins/speechmatics/__init__.py

@@ -30,7 +30,7 @@ from livekit.agents import Plugin
 
 
 class SpeechmaticsPlugin(Plugin):
-    def __init__(self):
+    def __init__(self) -> None:
         super().__init__(__name__, __version__, __package__)
 
 

livekit/plugins/speechmatics/stt.py

@@ -69,6 +69,7 @@ class STT(stt.STT):
                 operating_point="enhanced",
                 enable_partials=True,
                 max_delay=0.7,
+                speaker_diarization_config={"max_speakers": 2},
             )
         if not is_given(connection_settings):
             connection_settings = ConnectionSettings(  # noqa: B008
@@ -105,13 +106,13 @@ class STT(stt.STT):
         language: NotGivenOr[str] = NOT_GIVEN,
         conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS,
     ) -> SpeechStream:
-        config = dataclasses.replace(self._audio_settings)
+        transcription_config = dataclasses.replace(self._transcription_config)
         if is_given(language):
-            config.language = language
+            transcription_config.language = language
         stream = SpeechStream(
             stt=self,
-            transcription_config=self._transcription_config,
-            audio_settings=config,
+            transcription_config=transcription_config,
+            audio_settings=self._audio_settings,
             connection_settings=self._connection_settings,
             conn_options=conn_options,
             http_session=self.session,
@@ -145,10 +146,10 @@ class SpeechStream(stt.SpeechStream):
         self._recognition_started = asyncio.Event()
         self._seq_no = 0
 
-    async def _run(self):
+    async def _run(self) -> None:
         closing_ws = False
 
-        async def send_task(ws: aiohttp.ClientWebSocketResponse):
+        async def send_task(ws: aiohttp.ClientWebSocketResponse) -> None:
             nonlocal closing_ws
 
             start_recognition_msg = {
@@ -186,7 +187,7 @@ class SpeechStream(stt.SpeechStream):
                 )
             )
 
-        async def recv_task(ws: aiohttp.ClientWebSocketResponse):
+        async def recv_task(ws: aiohttp.ClientWebSocketResponse) -> None:
             nonlocal closing_ws
             while True:
                 msg = await ws.receive()
@@ -221,9 +222,9 @@ class SpeechStream(stt.SpeechStream):
 
                 try:
                     done, _ = await asyncio.wait(
-                        [tasks_group, wait_reconnect_task],
+                        (tasks_group, wait_reconnect_task),
                         return_when=asyncio.FIRST_COMPLETED,
-                    )  # type: ignore
+                    )
                     for task in done:
                         if task != wait_reconnect_task:
                             task.result()
@@ -316,6 +317,7 @@ def live_transcription_to_speech_data(data: dict) -> list[stt.SpeechData]:
                 alt.get("confidence", 1.0),
                 alt.get("language", "en"),
             )
+            speaker = alt.get("speaker")
 
             if not content:
                 continue
@@ -326,8 +328,9 @@ def live_transcription_to_speech_data(data: dict) -> list[stt.SpeechData]:
             elif speech_data and start_time == speech_data[-1].end_time:
                 speech_data[-1].text += " " + content
             else:
-                speech_data.append(
-                    stt.SpeechData(language, content, start_time, end_time, confidence)
-                )
+                sd = stt.SpeechData(language, content, start_time, end_time, confidence)
+                if speaker is not None:
+                    sd.speaker_id = speaker
+                speech_data.append(sd)
 
     return speech_data

livekit/plugins/speechmatics/types.py

@@ -49,6 +49,9 @@ class TranscriptionConfig:
     transcript_filtering_config: Optional[dict] = None
     """Removes disfluencies with the remove_disfluencies setting."""
 
+    speaker_diarization_config: Optional[dict] = None
+    """Options for speaker diarization such as ``max_speakers``."""
+
     def asdict(self) -> dict[Any, Any]:
         """Returns model as a dict while excluding None values recursively."""
         return asdict(self, dict_factory=lambda x: {k: v for (k, v) in x if v is not None})
@@ -65,7 +68,7 @@ class AudioSettings:
     sample_rate: int = 16000
     """Sampling rate in hertz."""
 
-    def asdict(self):
+    def asdict(self) -> dict[str, Any]:
         return {
             "type": "raw",
             "encoding": self.encoding,

livekit/plugins/speechmatics/utils.py

@@ -16,7 +16,7 @@ async def get_access_token(api_key: str) -> str:
             if resp.status == 201:
                 try:
                     data = await resp.json()
-                    return data["key_value"]
+                    return data["key_value"]  # type: ignore
                 except (ValueError, KeyError) as e:
                     raise Exception(
                         f"Failed to parse Speechmatics access token response: {e}"
@@ -29,12 +29,12 @@ async def get_access_token(api_key: str) -> str:
                 )
 
 
-def get_sdk_version():
+def get_sdk_version() -> str:
     version = importlib.metadata.version("livekit-plugins-speechmatics")
     return f"livekit-plugins-{version}"
 
 
-def sanitize_url(url, language):
+def sanitize_url(url: str, language: str) -> str:
     from urllib.parse import parse_qsl, urlencode, urlparse, urlunparse
 
     parsed_url = urlparse(url)

livekit/plugins/speechmatics/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.0.23"
+__version__ = "1.1.0"

{livekit_plugins_speechmatics-1.0.23.dist-info → livekit_plugins_speechmatics-1.1.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-speechmatics
-Version: 1.0.23
+Version: 1.1.0
 Summary: Agent Framework plugin for Speechmatics
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
@@ -18,7 +18,7 @@ Classifier: Topic :: Multimedia :: Sound/Audio
 Classifier: Topic :: Multimedia :: Video
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.9.0
-Requires-Dist: livekit-agents>=1.0.23
+Requires-Dist: livekit-agents>=1.1.0
 Description-Content-Type: text/markdown
 
 # Speechmatics STT plugin for LiveKit Agents

livekit_plugins_speechmatics-1.1.0.dist-info/RECORD

@@ -0,0 +1,10 @@
+livekit/plugins/speechmatics/__init__.py,sha256=LpY2NZfiSfEeqmjB37anBHdPMbNkSiNsnJ9GZ6Jf6Ac,1236
+livekit/plugins/speechmatics/log.py,sha256=O1iyAF7cHUu_iMXh6l7KRwwWeDB5QyABI_qzAb0cs04,75
+livekit/plugins/speechmatics/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+livekit/plugins/speechmatics/stt.py,sha256=10SRGs9j_8m0Enh3uw7tKvszNNErK90h4bD0mEHTO58,12000
+livekit/plugins/speechmatics/types.py,sha256=x9Jl5yABOU4hkN0adKEIhv0WLR32QWnVRo53qvyIPkY,4970
+livekit/plugins/speechmatics/utils.py,sha256=qVg7MenwUvp_f2AngntQlmZ4xb9b8c8NB5ftqGk1Wr0,1907
+livekit/plugins/speechmatics/version.py,sha256=lFVJxRoJYH4JZgg7iJQF0_dlrbfVbRYYWQzbHnv_p-w,600
+livekit_plugins_speechmatics-1.1.0.dist-info/METADATA,sha256=JaJtVnuE9UB6eFJj2wtdbiz64Qrwz8Q2lZk1xsPisEg,2049
+livekit_plugins_speechmatics-1.1.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_speechmatics-1.1.0.dist-info/RECORD,,

livekit_plugins_speechmatics-1.0.23.dist-info/RECORD

@@ -1,10 +0,0 @@
-livekit/plugins/speechmatics/__init__.py,sha256=mHlhJcJJc0aIhQxHx6fUJC9epfGjw1E8b-FdR0eUzMQ,1228
-livekit/plugins/speechmatics/log.py,sha256=O1iyAF7cHUu_iMXh6l7KRwwWeDB5QyABI_qzAb0cs04,75
-livekit/plugins/speechmatics/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-livekit/plugins/speechmatics/stt.py,sha256=NrK3cljwbx5dEgslepG3ROszvok-hZadanNHRNo7oFk,11775
-livekit/plugins/speechmatics/types.py,sha256=LG9ZL_r9f0LMmKPaWqCNzOU6wQJmXp8w7S23xuQaHO8,4829
-livekit/plugins/speechmatics/utils.py,sha256=9pbI8yrB8e06WY4NOH_RaoTSmY6JJc-P12-Sm8pW51U,1867
-livekit/plugins/speechmatics/version.py,sha256=O5JviWLJVBpC-DAakcGUwagiTFI5ZD7HTTTp2woaBmo,601
-livekit_plugins_speechmatics-1.0.23.dist-info/METADATA,sha256=MUNHq8MNBRHJkQCo8yBl42rfQrMxl5oocVxszMNPq14,2051
-livekit_plugins_speechmatics-1.0.23.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_speechmatics-1.0.23.dist-info/RECORD,,