cartesia 2.0.7__tar.gz → 2.0.9__tar.gz

{cartesia-2.0.7 → cartesia-2.0.9}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.3
+Metadata-Version: 2.1
 Name: cartesia
-Version: 2.0.7
+Version: 2.0.9
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers
@@ -16,7 +16,6 @@ Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
-Classifier: Programming Language :: Python :: 3.13
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Typing :: Typed
 Requires-Dist: aiohttp (>=3.10.10)
@@ -29,7 +28,6 @@ Requires-Dist: pydantic-core (>=2.18.2,<3.0.0)
 Requires-Dist: pydub (>=0.25.1)
 Requires-Dist: typing_extensions (>=4.0.0)
 Requires-Dist: websockets (>=10.4)
-Project-URL: Repository, https://github.com/cartesia-ai/cartesia-python
 Description-Content-Type: text/markdown
 
 # Cartesia Python Library

{cartesia-2.0.7 → cartesia-2.0.9}/pyproject.toml

@@ -3,7 +3,7 @@ name = "cartesia"
 
 [tool.poetry]
 name = "cartesia"
-version = "2.0.7"
+version = "2.0.9"
 description = ""
 readme = "README.md"
 authors = []

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/__init__.py

@@ -52,11 +52,7 @@ from .tts import (
     Controls,
     ControlsParams,
     Emotion,
-    ExperimentalModelControls,
-    ExperimentalModelControlsParams,
     FlushId,
-    GenerationConfig,
-    GenerationConfigParams,
     GenerationRequest,
     GenerationRequestParams,
     ModelSpeed,
@@ -217,16 +213,12 @@ __all__ = [
     "Emotion",
     "ErrorMessage",
     "ErrorMessageParams",
-    "ExperimentalModelControls",
-    "ExperimentalModelControlsParams",
     "FilePurpose",
     "FlushDoneMessage",
     "FlushDoneMessageParams",
     "FlushId",
     "Gender",
     "GenderPresentation",
-    "GenerationConfig",
-    "GenerationConfigParams",
     "GenerationRequest",
     "GenerationRequestParams",
     "GetVoicesResponse",

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/core/client_wrapper.py

@@ -16,7 +16,7 @@ class BaseClientWrapper:
         headers: typing.Dict[str, str] = {
             "X-Fern-Language": "Python",
             "X-Fern-SDK-Name": "cartesia",
-            "X-Fern-SDK-Version": "2.0.7",
+            "X-Fern-SDK-Version": "2.0.9",
         }
         headers["X-API-Key"] = self.api_key
         headers["Cartesia-Version"] = "2024-11-13"

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/__init__.py

@@ -5,9 +5,7 @@ from .types import (
     ContextId,
     Controls,
     Emotion,
-    ExperimentalModelControls,
     FlushId,
-    GenerationConfig,
     GenerationRequest,
     ModelSpeed,
     Mp3OutputFormat,
@@ -53,8 +51,6 @@ from .types import (
 from .requests import (
     CancelContextRequestParams,
     ControlsParams,
-    ExperimentalModelControlsParams,
-    GenerationConfigParams,
     GenerationRequestParams,
     Mp3OutputFormatParams,
     OutputFormatParams,
@@ -100,11 +96,7 @@ __all__ = [
     "Controls",
     "ControlsParams",
     "Emotion",
-    "ExperimentalModelControls",
-    "ExperimentalModelControlsParams",
     "FlushId",
-    "GenerationConfig",
-    "GenerationConfigParams",
     "GenerationRequest",
     "GenerationRequestParams",
     "ModelSpeed",

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/_async_websocket.py

@@ -50,7 +50,6 @@ class _AsyncTTSContext:
         self._context_id = context_id
         self._websocket = websocket
         self.timeout = timeout
-        self._error = None
 
     @property
     def context_id(self) -> str:
@@ -292,6 +291,7 @@ class AsyncTtsWebsocket(TtsWebsocket):
         self.websocket = None
         self._context_queues: Dict[str, List[asyncio.Queue]] = {}
         self._processing_task: Optional[asyncio.Task] = None
+        self._closing = False
 
     def __del__(self):
         try:
@@ -338,27 +338,28 @@ class AsyncTtsWebsocket(TtsWebsocket):
 
     async def close(self):
         """This method closes the websocket connection. *Highly* recommended to call this method when done."""
-        if self.websocket is not None and not self._is_websocket_closed():
-            await self.websocket.close()
-        if self._processing_task:
-            self._processing_task.cancel()
-            try:
-                self._processing_task = None
-            except asyncio.CancelledError:
-                pass
-            except TypeError as e:
-                # Ignore the error if the task is already canceled.
-                # For some reason we are getting None responses
-                # TODO: This needs to be fixed - we need to think about why we are getting None responses.
-                if "Received message 256:None" not in str(e):
-                    raise e
-
-        for context_id in list(self._context_queues.keys()):
-            self._remove_context(context_id)
-
-        self._context_queues.clear()
-        self._processing_task = None
-        self.websocket = None
+        self._closing = True
+        try:
+            if self._processing_task:
+                self._processing_task.cancel()
+                try:
+                    await self._processing_task
+                except asyncio.CancelledError:
+                    pass
+                finally:
+                    self._processing_task = None
+
+            if self.websocket is not None and not self._is_websocket_closed():
+                await self.websocket.close()
+
+            for context_id in list(self._context_queues.keys()):
+                self._remove_context(context_id)
+
+            self._context_queues.clear()
+            self._processing_task = None
+            self.websocket = None
+        finally:
+            self._closing = False
 
     async def send(
         self,
@@ -444,6 +445,16 @@ class AsyncTtsWebsocket(TtsWebsocket):
             ),
         )
 
+    def _is_close_error(self, e: TypeError) -> bool:
+        # TODO: This method checks the error string, but should check the code directly if/when aiohttp
+        # adds the code as a field to the error
+        exception_str = str(e)
+        return (
+            f"{aiohttp.WSMsgType.CLOSING}:None" in exception_str
+            or f"{aiohttp.WSMsgType.CLOSED}:None" in exception_str
+            or f"{aiohttp.WSMsgType.CLOSE}:1000" in exception_str
+        )
+
     async def _process_responses(self):
         try:
             while True:
@@ -453,8 +464,13 @@ class AsyncTtsWebsocket(TtsWebsocket):
                 flush_id = response.get("flush_id", -1)
                 if context_id in self._context_queues:
                     await self._context_queues[context_id][flush_id].put(response)
-        except Exception as e:
-            self._error = e
+        except TypeError as e:
+            # If the WebSocket is closing and the error is a close error, ignore it.
+            if self._closing and self._is_close_error(e):
+                return
+            # Otherwise, if this is a close error, raise a more clear error.
+            if self._is_close_error(e):
+                raise RuntimeError(f"WebSocket closed unexpectedly: {e}")
             raise e
 
     async def _get_message(

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/_websocket.py

@@ -46,7 +46,6 @@ class _TTSContext:
     def __init__(self, context_id: str, websocket: "TtsWebsocket"):
         self._context_id = context_id
         self._websocket = websocket
-        self._error = None
 
     def __del__(self):
         self._close()

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/client.py

@@ -6,7 +6,6 @@ from .requests.tts_request_voice_specifier import TtsRequestVoiceSpecifierParams
 from .requests.output_format import OutputFormatParams
 from .types.supported_language import SupportedLanguage
 from .types.model_speed import ModelSpeed
-from .requests.generation_config import GenerationConfigParams
 from ..core.request_options import RequestOptions
 from ..core.serialization import convert_and_respect_annotation_metadata
 from json.decoder import JSONDecodeError
@@ -37,7 +36,6 @@ class TtsClient:
         language: typing.Optional[SupportedLanguage] = OMIT,
         duration: typing.Optional[float] = OMIT,
         speed: typing.Optional[ModelSpeed] = OMIT,
-        generation_config: typing.Optional[GenerationConfigParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.Iterator[bytes]:
         """
@@ -60,8 +58,6 @@ class TtsClient:
 
         speed : typing.Optional[ModelSpeed]
 
-        generation_config : typing.Optional[GenerationConfigParams]
-
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration. You can pass in configuration such as `chunk_size`, and more to customize the request and response.
 
@@ -103,9 +99,6 @@ class TtsClient:
                 ),
                 "duration": duration,
                 "speed": speed,
-                "generation_config": convert_and_respect_annotation_metadata(
-                    object_=generation_config, annotation=GenerationConfigParams, direction="write"
-                ),
             },
             request_options=request_options,
             omit=OMIT,
@@ -257,7 +250,6 @@ class AsyncTtsClient:
         language: typing.Optional[SupportedLanguage] = OMIT,
         duration: typing.Optional[float] = OMIT,
         speed: typing.Optional[ModelSpeed] = OMIT,
-        generation_config: typing.Optional[GenerationConfigParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
     ) -> typing.AsyncIterator[bytes]:
         """
@@ -280,8 +272,6 @@ class AsyncTtsClient:
 
         speed : typing.Optional[ModelSpeed]
 
-        generation_config : typing.Optional[GenerationConfigParams]
-
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration. You can pass in configuration such as `chunk_size`, and more to customize the request and response.
 
@@ -331,9 +321,6 @@ class AsyncTtsClient:
                 ),
                 "duration": duration,
                 "speed": speed,
-                "generation_config": convert_and_respect_annotation_metadata(
-                    object_=generation_config, annotation=GenerationConfigParams, direction="write"
-                ),
             },
             request_options=request_options,
             omit=OMIT,

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/requests/__init__.py

@@ -2,8 +2,6 @@
 
 from .cancel_context_request import CancelContextRequestParams
 from .controls import ControlsParams
-from .experimental_model_controls import ExperimentalModelControlsParams
-from .generation_config import GenerationConfigParams
 from .generation_request import GenerationRequestParams
 from .mp_3_output_format import Mp3OutputFormatParams
 from .output_format import OutputFormatParams, OutputFormat_Mp3Params, OutputFormat_RawParams, OutputFormat_WavParams
@@ -43,8 +41,6 @@ from .word_timestamps import WordTimestampsParams
 __all__ = [
     "CancelContextRequestParams",
     "ControlsParams",
-    "ExperimentalModelControlsParams",
-    "GenerationConfigParams",
     "GenerationRequestParams",
     "Mp3OutputFormatParams",
     "OutputFormatParams",

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/requests/tts_request.py

@@ -6,7 +6,6 @@ import typing_extensions
 from ..types.supported_language import SupportedLanguage
 from .output_format import OutputFormatParams
 from ..types.model_speed import ModelSpeed
-from .generation_config import GenerationConfigParams
 
 
 class TtsRequestParams(typing_extensions.TypedDict):
@@ -26,4 +25,3 @@ class TtsRequestParams(typing_extensions.TypedDict):
     """
 
     speed: typing_extensions.NotRequired[ModelSpeed]
-    generation_config: typing_extensions.NotRequired[GenerationConfigParams]

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/types/__init__.py

@@ -4,9 +4,7 @@ from .cancel_context_request import CancelContextRequest
 from .context_id import ContextId
 from .controls import Controls
 from .emotion import Emotion
-from .experimental_model_controls import ExperimentalModelControls
 from .flush_id import FlushId
-from .generation_config import GenerationConfig
 from .generation_request import GenerationRequest
 from .model_speed import ModelSpeed
 from .mp_3_output_format import Mp3OutputFormat
@@ -53,9 +51,7 @@ __all__ = [
     "ContextId",
     "Controls",
     "Emotion",
-    "ExperimentalModelControls",
     "FlushId",
-    "GenerationConfig",
     "GenerationRequest",
     "ModelSpeed",
     "Mp3OutputFormat",

{cartesia-2.0.7 → cartesia-2.0.9}/src/cartesia/tts/types/tts_request.py

@@ -7,7 +7,6 @@ import typing
 from .supported_language import SupportedLanguage
 from .output_format import OutputFormat
 from .model_speed import ModelSpeed
-from .generation_config import GenerationConfig
 from ...core.pydantic_utilities import IS_PYDANTIC_V2
 
 
@@ -28,7 +27,6 @@ class TtsRequest(UniversalBaseModel):
     """
 
     speed: typing.Optional[ModelSpeed] = None
-    generation_config: typing.Optional[GenerationConfig] = None
 
     if IS_PYDANTIC_V2:
         model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2

cartesia-2.0.7/src/cartesia/tts/requests/experimental_model_controls.py

@@ -1,17 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing_extensions
-import typing_extensions
-
-
-class ExperimentalModelControlsParams(typing_extensions.TypedDict):
-    """
-    These controls are **experimental** and subject to breaking changes.
-    """
-
-    accent_localization: typing_extensions.NotRequired[int]
-    """
-    Toggle accent localization: 0 (disabled, default) or 1 (enabled).
-    When enabled, the voice adapts to match the transcript language's accent while preserving vocal characteristics. When disabled, maintains the original voice accent.
-    For more information, see [Localize Voices](/build-with-sonic/capabilities/localize-voices).
-    """

cartesia-2.0.7/src/cartesia/tts/requests/generation_config.py

@@ -1,23 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-import typing_extensions
-import typing_extensions
-from .experimental_model_controls import ExperimentalModelControlsParams
-
-
-class GenerationConfigParams(typing_extensions.TypedDict):
-    """
-    Configure the various attributes of the generated speech. These controls are only available for `sonic-3-preview` and will have no effect on earlier models.
-    """
-
-    volume: typing_extensions.NotRequired[float]
-    """
-    Adjust the volume of the generated speech between -1.0 (softer) and 1.0 (louder). 0.0 is the default volume.
-    """
-
-    speed: typing_extensions.NotRequired[float]
-    """
-    Adjust the speed of the generated speech between -1.0 (slower) and 1.0 (faster). 0.0 is the default speed.
-    """
-
-    experimental: typing_extensions.NotRequired[ExperimentalModelControlsParams]

cartesia-2.0.7/src/cartesia/tts/types/experimental_model_controls.py

@@ -1,28 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-from ...core.pydantic_utilities import UniversalBaseModel
-import typing
-import pydantic
-from ...core.pydantic_utilities import IS_PYDANTIC_V2
-
-
-class ExperimentalModelControls(UniversalBaseModel):
-    """
-    These controls are **experimental** and subject to breaking changes.
-    """
-
-    accent_localization: typing.Optional[int] = pydantic.Field(default=None)
-    """
-    Toggle accent localization: 0 (disabled, default) or 1 (enabled).
-    When enabled, the voice adapts to match the transcript language's accent while preserving vocal characteristics. When disabled, maintains the original voice accent.
-    For more information, see [Localize Voices](/build-with-sonic/capabilities/localize-voices).
-    """
-
-    if IS_PYDANTIC_V2:
-        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
-    else:
-
-        class Config:
-            frozen = True
-            smart_union = True
-            extra = pydantic.Extra.allow

cartesia-2.0.7/src/cartesia/tts/types/generation_config.py

@@ -1,34 +0,0 @@
-# This file was auto-generated by Fern from our API Definition.
-
-from ...core.pydantic_utilities import UniversalBaseModel
-import typing
-import pydantic
-from .experimental_model_controls import ExperimentalModelControls
-from ...core.pydantic_utilities import IS_PYDANTIC_V2
-
-
-class GenerationConfig(UniversalBaseModel):
-    """
-    Configure the various attributes of the generated speech. These controls are only available for `sonic-3-preview` and will have no effect on earlier models.
-    """
-
-    volume: typing.Optional[float] = pydantic.Field(default=None)
-    """
-    Adjust the volume of the generated speech between -1.0 (softer) and 1.0 (louder). 0.0 is the default volume.
-    """
-
-    speed: typing.Optional[float] = pydantic.Field(default=None)
-    """
-    Adjust the speed of the generated speech between -1.0 (slower) and 1.0 (faster). 0.0 is the default speed.
-    """
-
-    experimental: typing.Optional[ExperimentalModelControls] = None
-
-    if IS_PYDANTIC_V2:
-        model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2
-    else:
-
-        class Config:
-            frozen = True
-            smart_union = True
-            extra = pydantic.Extra.allow