livekit-plugins-aws 1.0.20__py3-none-any.whl → 1.0.22__py3-none-any.whl

livekit/plugins/aws/llm.py

@@ -31,7 +31,7 @@ from livekit.agents.types import (
 from livekit.agents.utils import is_given
 
 from .log import logger
-from .utils import get_aws_async_session, to_chat_ctx, to_fnc_ctx
+from .utils import to_chat_ctx, to_fnc_ctx
 
 TEXT_MODEL = Literal["anthropic.claude-3-5-sonnet-20241022-v2:0"]
 
@@ -83,10 +83,10 @@ class LLM(llm.LLM):
         """  # noqa: E501
         super().__init__()
 
-        self._session = session or get_aws_async_session(
-            api_key=api_key if is_given(api_key) else None,
-            api_secret=api_secret if is_given(api_secret) else None,
-            region=region if is_given(region) else None,
+        self._session = session or aioboto3.Session(
+            aws_access_key_id=api_key if is_given(api_key) else None,
+            aws_secret_access_key=api_secret if is_given(api_secret) else None,
+            region_name=region if is_given(region) else None,
         )
 
         model = model if is_given(model) else os.environ.get("BEDROCK_INFERENCE_PROFILE_ARN")

livekit/plugins/aws/stt.py

@@ -13,22 +13,26 @@
 from __future__ import annotations
 
 import asyncio
+import os
 from dataclasses import dataclass
 
-import aioboto3
-from amazon_transcribe.auth import StaticCredentialResolver
+from amazon_transcribe.auth import AwsCrtCredentialResolver
 from amazon_transcribe.client import TranscribeStreamingClient
-from amazon_transcribe.model import Result, TranscriptEvent
+from amazon_transcribe.exceptions import BadRequestException
+from amazon_transcribe.model import Result, StartStreamTranscriptionEventStream, TranscriptEvent
 
 from livekit import rtc
-from livekit.agents import DEFAULT_API_CONNECT_OPTIONS, APIConnectOptions, stt, utils
+from livekit.agents import (
+    DEFAULT_API_CONNECT_OPTIONS,
+    APIConnectOptions,
+    stt,
+    utils,
+)
 from livekit.agents.types import NOT_GIVEN, NotGivenOr
 from livekit.agents.utils import is_given
 
 from .log import logger
-from .utils import DEFAULT_REGION, get_aws_async_session
-
-REFRESH_INTERVAL = 1800
+from .utils import DEFAULT_REGION
 
 
 @dataclass
@@ -53,9 +57,7 @@ class STT(stt.STT):
         self,
         *,
         region: NotGivenOr[str] = NOT_GIVEN,
-        api_key: NotGivenOr[str] = NOT_GIVEN,
-        api_secret: NotGivenOr[str] = NOT_GIVEN,
-        sample_rate: int = 48000,
+        sample_rate: int = 24000,
         language: str = "en-US",
         encoding: str = "pcm",
         vocabulary_name: NotGivenOr[str] = NOT_GIVEN,
@@ -68,15 +70,15 @@ class STT(stt.STT):
         enable_partial_results_stabilization: NotGivenOr[bool] = NOT_GIVEN,
         partial_results_stability: NotGivenOr[str] = NOT_GIVEN,
         language_model_name: NotGivenOr[str] = NOT_GIVEN,
-        session: aioboto3.Session | None = None,
-        refresh_interval: NotGivenOr[int] = NOT_GIVEN,
     ):
         super().__init__(capabilities=stt.STTCapabilities(streaming=True, interim_results=True))
-        self._region = region if is_given(region) else DEFAULT_REGION
-        self._session = session or get_aws_async_session(
-            api_key=api_key if is_given(api_key) else None,
-            api_secret=api_secret if is_given(api_secret) else None,
+
+        if not is_given(region):
+            region = os.getenv("AWS_REGION") or DEFAULT_REGION
+        self._region = region
+        self._client = TranscribeStreamingClient(
             region=self._region,
+            credential_resolver=AwsCrtCredentialResolver(None),
         )
 
         self._config = STTOptions(
@@ -94,27 +96,8 @@ class STT(stt.STT):
             partial_results_stability=partial_results_stability,
             language_model_name=language_model_name,
         )
-        self._pool = utils.ConnectionPool[TranscribeStreamingClient](
-            connect_cb=self._create_client,
-            max_session_duration=refresh_interval
-            if is_given(refresh_interval)
-            else REFRESH_INTERVAL,
-        )
-
-    async def _create_client(self) -> TranscribeStreamingClient:
-        creds = await self._session.get_credentials()
-        frozen_credentials = await creds.get_frozen_credentials()
-        return TranscribeStreamingClient(
-            region=self._region,
-            credential_resolver=StaticCredentialResolver(
-                access_key_id=frozen_credentials.access_key,
-                secret_access_key=frozen_credentials.secret_key,
-                session_token=frozen_credentials.token,
-            ),
-        )
 
     async def aclose(self) -> None:
-        await self._pool.aclose()
         await super().aclose()
 
     async def _recognize_impl(
@@ -134,7 +117,7 @@ class STT(stt.STT):
     ) -> SpeechStream:
         return SpeechStream(
             stt=self,
-            pool=self._pool,
+            client=self._client,
             conn_options=conn_options,
             opts=self._config,
         )
@@ -145,15 +128,15 @@ class SpeechStream(stt.SpeechStream):
         self,
         stt: STT,
         opts: STTOptions,
-        pool: utils.ConnectionPool[TranscribeStreamingClient],
+        client: TranscribeStreamingClient,
         conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS,
     ) -> None:
         super().__init__(stt=stt, conn_options=conn_options, sample_rate=opts.sample_rate)
         self._opts = opts
-        self._pool = pool
+        self._client = client
 
     async def _run(self) -> None:
-        async with self._pool.connection() as client:
+        while True:
             live_config = {
                 "language_code": self._opts.language,
                 "media_sample_rate_hz": self._opts.sample_rate,
@@ -170,27 +153,34 @@ class SpeechStream(stt.SpeechStream):
                 "language_model_name": self._opts.language_model_name,
             }
             filtered_config = {k: v for k, v in live_config.items() if v and is_given(v)}
-            stream = await client.start_stream_transcription(**filtered_config)
+            stream = await self._client.start_stream_transcription(**filtered_config)
 
-            @utils.log_exceptions(logger=logger)
-            async def input_generator():
+            async def input_generator(stream: StartStreamTranscriptionEventStream):
                 async for frame in self._input_ch:
                     if isinstance(frame, rtc.AudioFrame):
                         await stream.input_stream.send_audio_event(audio_chunk=frame.data.tobytes())
                 await stream.input_stream.end_stream()
 
-            @utils.log_exceptions(logger=logger)
-            async def handle_transcript_events():
+            async def handle_transcript_events(stream: StartStreamTranscriptionEventStream):
                 async for event in stream.output_stream:
                     if isinstance(event, TranscriptEvent):
                         self._process_transcript_event(event)
 
             tasks = [
-                asyncio.create_task(input_generator()),
-                asyncio.create_task(handle_transcript_events()),
+                asyncio.create_task(input_generator(stream)),
+                asyncio.create_task(handle_transcript_events(stream)),
             ]
             try:
                 await asyncio.gather(*tasks)
+            except BadRequestException as e:
+                if e.message and e.message.startswith("Your request timed out"):
+                    # AWS times out after 15s of inactivity, this tends to happen
+                    # at the end of the session, when the input is gone, we'll ignore it and
+                    # just treat it as a silent retry
+                    logger.info("restarting transcribe session")
+                    continue
+                else:
+                    raise e
             finally:
                 await utils.aio.gracefully_cancel(*tasks)
 

livekit/plugins/aws/tts.py

@@ -34,7 +34,7 @@ from livekit.agents.types import (
 from livekit.agents.utils import is_given
 
 from .models import TTS_LANGUAGE, TTS_SPEECH_ENGINE
-from .utils import _strip_nones, get_aws_async_session
+from .utils import _strip_nones
 
 TTS_NUM_CHANNELS: int = 1
 DEFAULT_SPEECH_ENGINE: TTS_SPEECH_ENGINE = "generative"
@@ -90,10 +90,10 @@ class TTS(tts.TTS):
             sample_rate=sample_rate,
             num_channels=TTS_NUM_CHANNELS,
         )
-        self._session = session or get_aws_async_session(
-            api_key=api_key if is_given(api_key) else None,
-            api_secret=api_secret if is_given(api_secret) else None,
-            region=region if is_given(region) else None,
+        self._session = session or aioboto3.Session(
+            aws_access_key_id=api_key if is_given(api_key) else None,
+            aws_secret_access_key=api_secret if is_given(api_secret) else None,
+            region_name=region if is_given(region) else None,
         )
         self._opts = _TTSOptions(
             voice=voice,

livekit/plugins/aws/utils.py

@@ -3,44 +3,13 @@ from __future__ import annotations
 import json
 from typing import Any
 
-import aioboto3
-import boto3
-from botocore.exceptions import NoCredentialsError
-
 from livekit.agents import llm
 from livekit.agents.llm import ChatContext, FunctionTool, ImageContent, utils
 
-__all__ = ["to_fnc_ctx", "to_chat_ctx", "get_aws_async_session"]
+__all__ = ["to_fnc_ctx", "to_chat_ctx"]
 DEFAULT_REGION = "us-east-1"
 
 
-def get_aws_async_session(
-    region: str | None = None,
-    api_key: str | None = None,
-    api_secret: str | None = None,
-) -> aioboto3.Session:
-    _validate_aws_credentials(api_key, api_secret)
-    session = aioboto3.Session(
-        aws_access_key_id=api_key,
-        aws_secret_access_key=api_secret,
-        region_name=region or DEFAULT_REGION,
-    )
-    return session
-
-
-def _validate_aws_credentials(
-    api_key: str | None = None,
-    api_secret: str | None = None,
-) -> None:
-    try:
-        session = boto3.Session(aws_access_key_id=api_key, aws_secret_access_key=api_secret)
-        creds = session.get_credentials()
-        if not creds:
-            raise ValueError("No credentials found")
-    except (NoCredentialsError, Exception) as e:
-        raise ValueError(f"Unable to locate valid AWS credentials: {str(e)}") from e
-
-
 def to_fnc_ctx(fncs: list[FunctionTool]) -> list[dict]:
     return [_build_tool_spec(fnc) for fnc in fncs]
 

livekit/plugins/aws/version.py

@@ -12,4 +12,4 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-__version__ = "1.0.20"
+__version__ = "1.0.22"

{livekit_plugins_aws-1.0.20.dist-info → livekit_plugins_aws-1.0.22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: livekit-plugins-aws
-Version: 1.0.20
+Version: 1.0.22
 Summary: LiveKit Agents Plugin for services from AWS
 Project-URL: Documentation, https://docs.livekit.io
 Project-URL: Website, https://livekit.io/
@@ -18,10 +18,9 @@ Classifier: Topic :: Multimedia :: Sound/Audio
 Classifier: Topic :: Multimedia :: Video
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
 Requires-Python: >=3.9.0
-Requires-Dist: aioboto3==14.1.0
-Requires-Dist: amazon-transcribe==0.6.2
-Requires-Dist: boto3==1.37.1
-Requires-Dist: livekit-agents>=1.0.20
+Requires-Dist: aioboto3>=14.1.0
+Requires-Dist: amazon-transcribe>=0.6.2
+Requires-Dist: livekit-agents>=1.0.22
 Description-Content-Type: text/markdown
 
 # AWS plugin for LiveKit Agents

livekit_plugins_aws-1.0.22.dist-info/RECORD

@@ -0,0 +1,12 @@
+livekit/plugins/aws/__init__.py,sha256=fkbgTfNZc6z4VxbYGNdY73EoGvNuRcJiuD-OlUHvjHU,1322
+livekit/plugins/aws/llm.py,sha256=Z9FFjAZPKgX3NKE7J2rUBUVul-T-3sIHIl5W4b0bzLs,11249
+livekit/plugins/aws/log.py,sha256=jFief0Xhv0n_F6sp6UFu9VKxs2bXNVGAfYGmEYfR_2Q,66
+livekit/plugins/aws/models.py,sha256=Nf8RFmDulW7h03dG2lERTog3mgDK0TbLvW0eGOncuEE,704
+livekit/plugins/aws/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+livekit/plugins/aws/stt.py,sha256=JzCRxLec7ZbbNHUbjLl5PgzgvZBppeprQMmb_U5EpAI,8946
+livekit/plugins/aws/tts.py,sha256=iO4lBO1ggE2OBN5j54novJnwPa8GhyW0wiY-Xj8VDEk,7336
+livekit/plugins/aws/utils.py,sha256=UcKrSYrZf7SKYua1F51KZHskXRC5L5h-_WpFoPWLowo,4037
+livekit/plugins/aws/version.py,sha256=-8dkOE2vDSF9WN8VoBrSwU2sb5YBGFuwPnSQXQ-uaYM,601
+livekit_plugins_aws-1.0.22.dist-info/METADATA,sha256=Que989diZatTKKrH49PHZnNtletiCi6nt7uxJtPqT_8,1531
+livekit_plugins_aws-1.0.22.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+livekit_plugins_aws-1.0.22.dist-info/RECORD,,

livekit_plugins_aws-1.0.20.dist-info/RECORD

@@ -1,12 +0,0 @@
-livekit/plugins/aws/__init__.py,sha256=fkbgTfNZc6z4VxbYGNdY73EoGvNuRcJiuD-OlUHvjHU,1322
-livekit/plugins/aws/llm.py,sha256=k9RTldvqg3uFW36jjBrrT0pdMyJJH_88RwI--0f10Pw,11251
-livekit/plugins/aws/log.py,sha256=jFief0Xhv0n_F6sp6UFu9VKxs2bXNVGAfYGmEYfR_2Q,66
-livekit/plugins/aws/models.py,sha256=Nf8RFmDulW7h03dG2lERTog3mgDK0TbLvW0eGOncuEE,704
-livekit/plugins/aws/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-livekit/plugins/aws/stt.py,sha256=FgZ0hN-ToqQRzesdV28GNkHw_x2X_sFugji_BnLXm7c,9448
-livekit/plugins/aws/tts.py,sha256=-YK_jgJxQUG5LKHNeqTfjOm6fQhoEEbHps300goTtMQ,7338
-livekit/plugins/aws/utils.py,sha256=BqZPyLr-xETbGybhE3-lEJovqkuCuAd-cxtUO3aFAVM,4988
-livekit/plugins/aws/version.py,sha256=t4KmPVTpEy1pOJ2GRCA-GNJfCQq_-zHNDBxGj4GKfVk,601
-livekit_plugins_aws-1.0.20.dist-info/METADATA,sha256=JThAx7JyHInOgQHgs92i1BKypGnYZQCWa11rE2Eayg0,1560
-livekit_plugins_aws-1.0.20.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-livekit_plugins_aws-1.0.20.dist-info/RECORD,,