cartesia 2.0.0b7__py3-none-any.whl → 2.0.0b8__py3-none-any.whl

cartesia/__init__.py

@@ -129,6 +129,9 @@ from .voices import (
     LocalizeDialect,
     LocalizeDialectParams,
     LocalizeEnglishDialect,
+    LocalizeFrenchDialect,
+    LocalizePortugueseDialect,
+    LocalizeSpanishDialect,
     LocalizeTargetLanguage,
     LocalizeVoiceRequest,
     LocalizeVoiceRequestParams,
@@ -187,6 +190,9 @@ __all__ = [
     "LocalizeDialect",
     "LocalizeDialectParams",
     "LocalizeEnglishDialect",
+    "LocalizeFrenchDialect",
+    "LocalizePortugueseDialect",
+    "LocalizeSpanishDialect",
     "LocalizeTargetLanguage",
     "LocalizeVoiceRequest",
     "LocalizeVoiceRequestParams",

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": "2.0.0b7",
+            "X-Fern-SDK-Version": "2.0.0b8",
         }
         headers["X-API-Key"] = self.api_key
         headers["Cartesia-Version"] = "2024-11-13"

cartesia/tts/_async_websocket.py

@@ -69,6 +69,7 @@ class _AsyncTTSContext:
         stream: bool = True,
         add_timestamps: bool = False,
         add_phoneme_timestamps: bool = False,
+        use_original_timestamps: bool = False,
         continue_: bool = False,
         flush: bool = False,
     ) -> None:
@@ -106,6 +107,8 @@ class _AsyncTTSContext:
             request_body["add_timestamps"] = add_timestamps
         if add_phoneme_timestamps:
             request_body["add_phoneme_timestamps"] = add_phoneme_timestamps
+        if use_original_timestamps:
+            request_body["use_original_timestamps"] = use_original_timestamps
         if continue_:
             request_body["continue"] = continue_
         if flush:
@@ -367,6 +370,7 @@ class AsyncTtsWebsocket(TtsWebsocket):
         stream: bool = True,
         add_timestamps: bool = False,
         add_phoneme_timestamps: bool = False,
+        use_original_timestamps: bool = False,
     ):
         """See :meth:`_WebSocket.send` for details."""
         if context_id is None:
@@ -385,6 +389,7 @@ class AsyncTtsWebsocket(TtsWebsocket):
             continue_=False,
             add_timestamps=add_timestamps,
             add_phoneme_timestamps=add_phoneme_timestamps,
+            use_original_timestamps=use_original_timestamps,
         )
 
         generator = ctx.receive()

cartesia/tts/_websocket.py

@@ -67,6 +67,8 @@ class _TTSContext:
         language: Optional[str] = None,
         stream: bool = True,
         add_timestamps: bool = False,
+        add_phoneme_timestamps: bool = False,
+        use_original_timestamps: bool = False,
     ) -> Generator[bytes, None, None]:
         """Send audio generation requests to the WebSocket and yield responses.
 
@@ -102,6 +104,10 @@ class _TTSContext:
             request_body["stream"] = stream
         if add_timestamps:
             request_body["add_timestamps"] = add_timestamps
+        if add_phoneme_timestamps:
+            request_body["add_phoneme_timestamps"] = add_phoneme_timestamps
+        if use_original_timestamps:
+            request_body["use_original_timestamps"] = use_original_timestamps
 
         if (
             "context_id" in request_body
@@ -354,6 +360,7 @@ class TtsWebsocket:
         stream: bool = True,
         add_timestamps: bool = False,
         add_phoneme_timestamps: bool = False,
+        use_original_timestamps: bool = False,
     ):
         """Send a request to the WebSocket to generate audio.
 
@@ -384,6 +391,7 @@ class TtsWebsocket:
             "stream": stream,
             "add_timestamps": add_timestamps,
             "add_phoneme_timestamps": add_phoneme_timestamps,
+            "use_original_timestamps": use_original_timestamps,
         }
         generator = self._websocket_generator(request_body)
 

cartesia/voices/__init__.py

@@ -12,6 +12,9 @@ from .types import (
     IdSpecifier,
     LocalizeDialect,
     LocalizeEnglishDialect,
+    LocalizeFrenchDialect,
+    LocalizePortugueseDialect,
+    LocalizeSpanishDialect,
     LocalizeTargetLanguage,
     LocalizeVoiceRequest,
     MixVoiceSpecifier,
@@ -56,6 +59,9 @@ __all__ = [
     "LocalizeDialect",
     "LocalizeDialectParams",
     "LocalizeEnglishDialect",
+    "LocalizeFrenchDialect",
+    "LocalizePortugueseDialect",
+    "LocalizeSpanishDialect",
     "LocalizeTargetLanguage",
     "LocalizeVoiceRequest",
     "LocalizeVoiceRequestParams",

cartesia/voices/client.py

@@ -11,19 +11,20 @@ from .types.get_voices_response import GetVoicesResponse
 from ..core.pydantic_utilities import parse_obj_as
 from json.decoder import JSONDecodeError
 from ..core.api_error import ApiError
-from ..embedding.types.embedding import Embedding
+from .. import core
 from ..tts.types.supported_language import SupportedLanguage
+from .types.clone_mode import CloneMode
+from .types.voice_metadata import VoiceMetadata
 from .types.voice_id import VoiceId
 from ..core.jsonable_encoder import jsonable_encoder
 from .types.localize_target_language import LocalizeTargetLanguage
 from .types.gender import Gender
 from .requests.localize_dialect import LocalizeDialectParams
-from .types.embedding_response import EmbeddingResponse
 from ..core.serialization import convert_and_respect_annotation_metadata
 from .requests.mix_voice_specifier import MixVoiceSpecifierParams
-from .. import core
-from .types.clone_mode import CloneMode
-from .types.voice_metadata import VoiceMetadata
+from .types.embedding_response import EmbeddingResponse
+from ..embedding.types.embedding import Embedding
+from .types.base_voice_id import BaseVoiceId
 from ..core.client_wrapper import AsyncClientWrapper
 from ..core.pagination import AsyncPager
 
@@ -140,34 +141,60 @@ class VoicesClient:
             raise ApiError(status_code=_response.status_code, body=_response.text)
         raise ApiError(status_code=_response.status_code, body=_response_json)
 
-    def create(
+    def clone(
         self,
         *,
+        clip: core.File,
         name: str,
-        description: str,
-        embedding: Embedding,
-        language: typing.Optional[SupportedLanguage] = OMIT,
+        language: SupportedLanguage,
+        mode: CloneMode,
+        enhance: bool,
+        description: typing.Optional[str] = OMIT,
+        transcript: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> Voice:
+    ) -> VoiceMetadata:
         """
+        Clone a voice from an audio clip. This endpoint has two modes, stability and similarity.
+
+        Similarity mode clones are more similar to the source clip, but may reproduce background noise. For these, use an audio clip about 5 seconds long.
+
+        Stability mode clones are more stable, but may not sound as similar to the source clip. For these, use an audio clip 10-20 seconds long.
+
         Parameters
         ----------
+        clip : core.File
+            See core.File for more documentation
+
         name : str
             The name of the voice.
 
-        description : str
-            The description of the voice.
 
-        embedding : Embedding
+        language : SupportedLanguage
+            The language of the voice.
+
+
+        mode : CloneMode
+            Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.
+
+
+        enhance : bool
+            Whether to enhance the clip to improve its quality before cloning. Useful if the clip has background noise.
+
+
+        description : typing.Optional[str]
+            A description for the voice.
+
+
+        transcript : typing.Optional[str]
+            Optional transcript of the words spoken in the audio clip. Only used for similarity mode.
 
-        language : typing.Optional[SupportedLanguage]
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
         Returns
         -------
-        Voice
+        VoiceMetadata
 
         Examples
         --------
@@ -176,20 +203,27 @@ class VoicesClient:
         client = Cartesia(
             api_key="YOUR_API_KEY",
         )
-        client.voices.create(
-            name="name",
-            description="description",
-            embedding=[1.1, 1.1],
+        client.voices.clone(
+            name="A high-stability cloned voice",
+            description="Copied from Cartesia docs",
+            mode="stability",
+            language="en",
+            enhance=True,
         )
         """
         _response = self._client_wrapper.httpx_client.request(
-            "voices/",
+            "voices/clone",
             method="POST",
-            json={
+            data={
                 "name": name,
                 "description": description,
-                "embedding": embedding,
                 "language": language,
+                "mode": mode,
+                "enhance": enhance,
+                "transcript": transcript,
+            },
+            files={
+                "clip": clip,
             },
             request_options=request_options,
             omit=OMIT,
@@ -197,9 +231,9 @@ class VoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    Voice,
+                    VoiceMetadata,
                     parse_obj_as(
-                        type_=Voice,  # type: ignore
+                        type_=VoiceMetadata,  # type: ignore
                         object_=_response.json(),
                     ),
                 )
@@ -349,16 +383,27 @@ class VoicesClient:
     def localize(
         self,
         *,
-        embedding: Embedding,
+        voice_id: str,
+        name: str,
+        description: str,
         language: LocalizeTargetLanguage,
         original_speaker_gender: Gender,
         dialect: typing.Optional[LocalizeDialectParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> EmbeddingResponse:
+    ) -> VoiceMetadata:
         """
+        Create a new voice from an existing voice localized to a new language and dialect.
+
         Parameters
         ----------
-        embedding : Embedding
+        voice_id : str
+            The ID of the voice to localize.
+
+        name : str
+            The name of the new localized voice.
+
+        description : str
+            The description of the new localized voice.
 
         language : LocalizeTargetLanguage
 
@@ -371,7 +416,7 @@ class VoicesClient:
 
         Returns
         -------
-        EmbeddingResponse
+        VoiceMetadata
 
         Examples
         --------
@@ -381,16 +426,21 @@ class VoicesClient:
             api_key="YOUR_API_KEY",
         )
         client.voices.localize(
-            embedding=[1.1, 1.1],
-            language="en",
-            original_speaker_gender="male",
+            voice_id="694f9389-aac1-45b6-b726-9d9369183238",
+            name="Sarah Peninsular Spanish",
+            description="Sarah Voice in Peninsular Spanish",
+            language="es",
+            original_speaker_gender="female",
+            dialect="pe",
         )
         """
         _response = self._client_wrapper.httpx_client.request(
             "voices/localize",
             method="POST",
             json={
-                "embedding": embedding,
+                "voice_id": voice_id,
+                "name": name,
+                "description": description,
                 "language": language,
                 "original_speaker_gender": original_speaker_gender,
                 "dialect": convert_and_respect_annotation_metadata(
@@ -403,9 +453,9 @@ class VoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    EmbeddingResponse,
+                    VoiceMetadata,
                     parse_obj_as(
-                        type_=EmbeddingResponse,  # type: ignore
+                        type_=VoiceMetadata,  # type: ignore
                         object_=_response.json(),
                     ),
                 )
@@ -468,58 +518,39 @@ class VoicesClient:
             raise ApiError(status_code=_response.status_code, body=_response.text)
         raise ApiError(status_code=_response.status_code, body=_response_json)
 
-    def clone(
+    def create(
         self,
         *,
-        clip: core.File,
         name: str,
-        language: SupportedLanguage,
-        mode: CloneMode,
-        enhance: bool,
-        description: typing.Optional[str] = OMIT,
-        transcript: typing.Optional[str] = OMIT,
+        description: str,
+        embedding: Embedding,
+        language: typing.Optional[SupportedLanguage] = OMIT,
+        base_voice_id: typing.Optional[BaseVoiceId] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> VoiceMetadata:
+    ) -> Voice:
         """
-        Clone a voice from an audio clip. This endpoint has two modes, stability and similarity.
-        Similarity mode clones are more similar to the source clip, but may reproduce background noise. For these, use an audio clip about 5 seconds long.
-        Stability mode clones are more stable, but may not sound as similar to the source clip. For these, use an audio clip 10-20 seconds long.
+        Create voice from raw features. If you'd like to clone a voice from an audio file, please use Clone Voice instead.
 
         Parameters
         ----------
-        clip : core.File
-            See core.File for more documentation
-
         name : str
             The name of the voice.
 
+        description : str
+            The description of the voice.
 
-        language : SupportedLanguage
-            The language of the voice.
-
-
-        mode : CloneMode
-            Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.
-
-
-        enhance : bool
-            Whether to enhance the clip to improve its quality before cloning. Useful if the clip has background noise.
-
-
-        description : typing.Optional[str]
-            A description for the voice.
-
+        embedding : Embedding
 
-        transcript : typing.Optional[str]
-            Optional transcript of the words spoken in the audio clip. Only used for similarity mode.
+        language : typing.Optional[SupportedLanguage]
 
+        base_voice_id : typing.Optional[BaseVoiceId]
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
         Returns
         -------
-        VoiceMetadata
+        Voice
 
         Examples
         --------
@@ -528,27 +559,21 @@ class VoicesClient:
         client = Cartesia(
             api_key="YOUR_API_KEY",
         )
-        client.voices.clone(
-            name="A high-stability cloned voice",
-            description="Copied from Cartesia docs",
-            mode="stability",
-            language="en",
-            enhance=True,
+        client.voices.create(
+            name="name",
+            description="description",
+            embedding=[1.1, 1.1],
         )
         """
         _response = self._client_wrapper.httpx_client.request(
-            "voices/clone",
+            "voices/",
             method="POST",
-            data={
+            json={
                 "name": name,
                 "description": description,
+                "embedding": embedding,
                 "language": language,
-                "mode": mode,
-                "enhance": enhance,
-                "transcript": transcript,
-            },
-            files={
-                "clip": clip,
+                "base_voice_id": base_voice_id,
             },
             request_options=request_options,
             omit=OMIT,
@@ -556,9 +581,9 @@ class VoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    VoiceMetadata,
+                    Voice,
                     parse_obj_as(
-                        type_=VoiceMetadata,  # type: ignore
+                        type_=Voice,  # type: ignore
                         object_=_response.json(),
                     ),
                 )
@@ -685,34 +710,60 @@ class AsyncVoicesClient:
             raise ApiError(status_code=_response.status_code, body=_response.text)
         raise ApiError(status_code=_response.status_code, body=_response_json)
 
-    async def create(
+    async def clone(
         self,
         *,
+        clip: core.File,
         name: str,
-        description: str,
-        embedding: Embedding,
-        language: typing.Optional[SupportedLanguage] = OMIT,
+        language: SupportedLanguage,
+        mode: CloneMode,
+        enhance: bool,
+        description: typing.Optional[str] = OMIT,
+        transcript: typing.Optional[str] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> Voice:
+    ) -> VoiceMetadata:
         """
+        Clone a voice from an audio clip. This endpoint has two modes, stability and similarity.
+
+        Similarity mode clones are more similar to the source clip, but may reproduce background noise. For these, use an audio clip about 5 seconds long.
+
+        Stability mode clones are more stable, but may not sound as similar to the source clip. For these, use an audio clip 10-20 seconds long.
+
         Parameters
         ----------
+        clip : core.File
+            See core.File for more documentation
+
         name : str
             The name of the voice.
 
-        description : str
-            The description of the voice.
 
-        embedding : Embedding
+        language : SupportedLanguage
+            The language of the voice.
+
+
+        mode : CloneMode
+            Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.
+
+
+        enhance : bool
+            Whether to enhance the clip to improve its quality before cloning. Useful if the clip has background noise.
+
+
+        description : typing.Optional[str]
+            A description for the voice.
+
+
+        transcript : typing.Optional[str]
+            Optional transcript of the words spoken in the audio clip. Only used for similarity mode.
 
-        language : typing.Optional[SupportedLanguage]
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
         Returns
         -------
-        Voice
+        VoiceMetadata
 
         Examples
         --------
@@ -726,23 +777,30 @@ class AsyncVoicesClient:
 
 
         async def main() -> None:
-            await client.voices.create(
-                name="name",
-                description="description",
-                embedding=[1.1, 1.1],
+            await client.voices.clone(
+                name="A high-stability cloned voice",
+                description="Copied from Cartesia docs",
+                mode="stability",
+                language="en",
+                enhance=True,
             )
 
 
         asyncio.run(main())
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "voices/",
+            "voices/clone",
             method="POST",
-            json={
+            data={
                 "name": name,
                 "description": description,
-                "embedding": embedding,
                 "language": language,
+                "mode": mode,
+                "enhance": enhance,
+                "transcript": transcript,
+            },
+            files={
+                "clip": clip,
             },
             request_options=request_options,
             omit=OMIT,
@@ -750,9 +808,9 @@ class AsyncVoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    Voice,
+                    VoiceMetadata,
                     parse_obj_as(
-                        type_=Voice,  # type: ignore
+                        type_=VoiceMetadata,  # type: ignore
                         object_=_response.json(),
                     ),
                 )
@@ -926,16 +984,27 @@ class AsyncVoicesClient:
     async def localize(
         self,
         *,
-        embedding: Embedding,
+        voice_id: str,
+        name: str,
+        description: str,
         language: LocalizeTargetLanguage,
         original_speaker_gender: Gender,
         dialect: typing.Optional[LocalizeDialectParams] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> EmbeddingResponse:
+    ) -> VoiceMetadata:
         """
+        Create a new voice from an existing voice localized to a new language and dialect.
+
         Parameters
         ----------
-        embedding : Embedding
+        voice_id : str
+            The ID of the voice to localize.
+
+        name : str
+            The name of the new localized voice.
+
+        description : str
+            The description of the new localized voice.
 
         language : LocalizeTargetLanguage
 
@@ -948,7 +1017,7 @@ class AsyncVoicesClient:
 
         Returns
         -------
-        EmbeddingResponse
+        VoiceMetadata
 
         Examples
         --------
@@ -963,9 +1032,12 @@ class AsyncVoicesClient:
 
         async def main() -> None:
             await client.voices.localize(
-                embedding=[1.1, 1.1],
-                language="en",
-                original_speaker_gender="male",
+                voice_id="694f9389-aac1-45b6-b726-9d9369183238",
+                name="Sarah Peninsular Spanish",
+                description="Sarah Voice in Peninsular Spanish",
+                language="es",
+                original_speaker_gender="female",
+                dialect="pe",
             )
 
 
@@ -975,7 +1047,9 @@ class AsyncVoicesClient:
             "voices/localize",
             method="POST",
             json={
-                "embedding": embedding,
+                "voice_id": voice_id,
+                "name": name,
+                "description": description,
                 "language": language,
                 "original_speaker_gender": original_speaker_gender,
                 "dialect": convert_and_respect_annotation_metadata(
@@ -988,9 +1062,9 @@ class AsyncVoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    EmbeddingResponse,
+                    VoiceMetadata,
                     parse_obj_as(
-                        type_=EmbeddingResponse,  # type: ignore
+                        type_=VoiceMetadata,  # type: ignore
                         object_=_response.json(),
                     ),
                 )
@@ -1061,58 +1135,39 @@ class AsyncVoicesClient:
             raise ApiError(status_code=_response.status_code, body=_response.text)
         raise ApiError(status_code=_response.status_code, body=_response_json)
 
-    async def clone(
+    async def create(
         self,
         *,
-        clip: core.File,
         name: str,
-        language: SupportedLanguage,
-        mode: CloneMode,
-        enhance: bool,
-        description: typing.Optional[str] = OMIT,
-        transcript: typing.Optional[str] = OMIT,
+        description: str,
+        embedding: Embedding,
+        language: typing.Optional[SupportedLanguage] = OMIT,
+        base_voice_id: typing.Optional[BaseVoiceId] = OMIT,
         request_options: typing.Optional[RequestOptions] = None,
-    ) -> VoiceMetadata:
+    ) -> Voice:
         """
-        Clone a voice from an audio clip. This endpoint has two modes, stability and similarity.
-        Similarity mode clones are more similar to the source clip, but may reproduce background noise. For these, use an audio clip about 5 seconds long.
-        Stability mode clones are more stable, but may not sound as similar to the source clip. For these, use an audio clip 10-20 seconds long.
+        Create voice from raw features. If you'd like to clone a voice from an audio file, please use Clone Voice instead.
 
         Parameters
         ----------
-        clip : core.File
-            See core.File for more documentation
-
         name : str
             The name of the voice.
 
+        description : str
+            The description of the voice.
 
-        language : SupportedLanguage
-            The language of the voice.
-
-
-        mode : CloneMode
-            Tradeoff between similarity and stability. Similarity clones sound more like the source clip, but may reproduce background noise. Stability clones always sound like a studio recording, but may not sound as similar to the source clip.
-
-
-        enhance : bool
-            Whether to enhance the clip to improve its quality before cloning. Useful if the clip has background noise.
-
-
-        description : typing.Optional[str]
-            A description for the voice.
-
+        embedding : Embedding
 
-        transcript : typing.Optional[str]
-            Optional transcript of the words spoken in the audio clip. Only used for similarity mode.
+        language : typing.Optional[SupportedLanguage]
 
+        base_voice_id : typing.Optional[BaseVoiceId]
 
         request_options : typing.Optional[RequestOptions]
             Request-specific configuration.
 
         Returns
         -------
-        VoiceMetadata
+        Voice
 
         Examples
         --------
@@ -1126,30 +1181,24 @@ class AsyncVoicesClient:
 
 
         async def main() -> None:
-            await client.voices.clone(
-                name="A high-stability cloned voice",
-                description="Copied from Cartesia docs",
-                mode="stability",
-                language="en",
-                enhance=True,
+            await client.voices.create(
+                name="name",
+                description="description",
+                embedding=[1.1, 1.1],
             )
 
 
         asyncio.run(main())
         """
         _response = await self._client_wrapper.httpx_client.request(
-            "voices/clone",
+            "voices/",
             method="POST",
-            data={
+            json={
                 "name": name,
                 "description": description,
+                "embedding": embedding,
                 "language": language,
-                "mode": mode,
-                "enhance": enhance,
-                "transcript": transcript,
-            },
-            files={
-                "clip": clip,
+                "base_voice_id": base_voice_id,
             },
             request_options=request_options,
             omit=OMIT,
@@ -1157,9 +1206,9 @@ class AsyncVoicesClient:
         try:
             if 200 <= _response.status_code < 300:
                 return typing.cast(
-                    VoiceMetadata,
+                    Voice,
                     parse_obj_as(
-                        type_=VoiceMetadata,  # type: ignore
+                        type_=Voice,  # type: ignore
                         object_=_response.json(),
                     ),
                 )

cartesia/voices/requests/create_voice_request.py

@@ -4,6 +4,7 @@ import typing_extensions
 from ...embedding.types.embedding import Embedding
 import typing_extensions
 from ...tts.types.supported_language import SupportedLanguage
+from ..types.base_voice_id import BaseVoiceId
 
 
 class CreateVoiceRequestParams(typing_extensions.TypedDict):
@@ -19,3 +20,4 @@ class CreateVoiceRequestParams(typing_extensions.TypedDict):
 
     embedding: Embedding
     language: typing_extensions.NotRequired[SupportedLanguage]
+    base_voice_id: typing_extensions.NotRequired[BaseVoiceId]

cartesia/voices/requests/localize_dialect.py

@@ -2,5 +2,10 @@
 
 import typing
 from ..types.localize_english_dialect import LocalizeEnglishDialect
+from ..types.localize_spanish_dialect import LocalizeSpanishDialect
+from ..types.localize_portuguese_dialect import LocalizePortugueseDialect
+from ..types.localize_french_dialect import LocalizeFrenchDialect
 
-LocalizeDialectParams = typing.Union[LocalizeEnglishDialect]
+LocalizeDialectParams = typing.Union[
+    LocalizeEnglishDialect, LocalizeSpanishDialect, LocalizePortugueseDialect, LocalizeFrenchDialect
+]

cartesia/voices/requests/localize_voice_request.py

@@ -1,7 +1,6 @@
 # This file was auto-generated by Fern from our API Definition.
 
 import typing_extensions
-from ...embedding.types.embedding import Embedding
 from ..types.localize_target_language import LocalizeTargetLanguage
 from ..types.gender import Gender
 import typing_extensions
@@ -9,7 +8,21 @@ from .localize_dialect import LocalizeDialectParams
 
 
 class LocalizeVoiceRequestParams(typing_extensions.TypedDict):
-    embedding: Embedding
+    voice_id: str
+    """
+    The ID of the voice to localize.
+    """
+
+    name: str
+    """
+    The name of the new localized voice.
+    """
+
+    description: str
+    """
+    The description of the new localized voice.
+    """
+
     language: LocalizeTargetLanguage
     original_speaker_gender: Gender
     dialect: typing_extensions.NotRequired[LocalizeDialectParams]

cartesia/voices/types/__init__.py

@@ -11,6 +11,9 @@ from .get_voices_response import GetVoicesResponse
 from .id_specifier import IdSpecifier
 from .localize_dialect import LocalizeDialect
 from .localize_english_dialect import LocalizeEnglishDialect
+from .localize_french_dialect import LocalizeFrenchDialect
+from .localize_portuguese_dialect import LocalizePortugueseDialect
+from .localize_spanish_dialect import LocalizeSpanishDialect
 from .localize_target_language import LocalizeTargetLanguage
 from .localize_voice_request import LocalizeVoiceRequest
 from .mix_voice_specifier import MixVoiceSpecifier
@@ -34,6 +37,9 @@ __all__ = [
     "IdSpecifier",
     "LocalizeDialect",
     "LocalizeEnglishDialect",
+    "LocalizeFrenchDialect",
+    "LocalizePortugueseDialect",
+    "LocalizeSpanishDialect",
     "LocalizeTargetLanguage",
     "LocalizeVoiceRequest",
     "MixVoiceSpecifier",

cartesia/voices/types/create_voice_request.py

@@ -5,6 +5,7 @@ import pydantic
 from ...embedding.types.embedding import Embedding
 import typing
 from ...tts.types.supported_language import SupportedLanguage
+from .base_voice_id import BaseVoiceId
 from ...core.pydantic_utilities import IS_PYDANTIC_V2
 
 
@@ -21,6 +22,7 @@ class CreateVoiceRequest(UniversalBaseModel):
 
     embedding: Embedding
     language: typing.Optional[SupportedLanguage] = None
+    base_voice_id: typing.Optional[BaseVoiceId] = None
 
     if IS_PYDANTIC_V2:
         model_config: typing.ClassVar[pydantic.ConfigDict] = pydantic.ConfigDict(extra="allow", frozen=True)  # type: ignore # Pydantic v2

cartesia/voices/types/localize_dialect.py

@@ -2,5 +2,10 @@
 
 import typing
 from .localize_english_dialect import LocalizeEnglishDialect
+from .localize_spanish_dialect import LocalizeSpanishDialect
+from .localize_portuguese_dialect import LocalizePortugueseDialect
+from .localize_french_dialect import LocalizeFrenchDialect
 
-LocalizeDialect = typing.Union[LocalizeEnglishDialect]
+LocalizeDialect = typing.Union[
+    LocalizeEnglishDialect, LocalizeSpanishDialect, LocalizePortugueseDialect, LocalizeFrenchDialect
+]

cartesia/voices/types/localize_french_dialect.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+LocalizeFrenchDialect = typing.Union[typing.Literal["eu", "ca"], typing.Any]

cartesia/voices/types/localize_portuguese_dialect.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+LocalizePortugueseDialect = typing.Union[typing.Literal["br", "eu"], typing.Any]

cartesia/voices/types/localize_spanish_dialect.py

@@ -0,0 +1,5 @@
+# This file was auto-generated by Fern from our API Definition.
+
+import typing
+
+LocalizeSpanishDialect = typing.Union[typing.Literal["mx", "pe"], typing.Any]

cartesia/voices/types/localize_voice_request.py

@@ -1,17 +1,30 @@
 # This file was auto-generated by Fern from our API Definition.
 
 from ...core.pydantic_utilities import UniversalBaseModel
-from ...embedding.types.embedding import Embedding
+import pydantic
 from .localize_target_language import LocalizeTargetLanguage
 from .gender import Gender
 import typing
 from .localize_dialect import LocalizeDialect
 from ...core.pydantic_utilities import IS_PYDANTIC_V2
-import pydantic
 
 
 class LocalizeVoiceRequest(UniversalBaseModel):
-    embedding: Embedding
+    voice_id: str = pydantic.Field()
+    """
+    The ID of the voice to localize.
+    """
+
+    name: str = pydantic.Field()
+    """
+    The name of the new localized voice.
+    """
+
+    description: str = pydantic.Field()
+    """
+    The description of the new localized voice.
+    """
+
     language: LocalizeTargetLanguage
     original_speaker_gender: Gender
     dialect: typing.Optional[LocalizeDialect] = None

{cartesia-2.0.0b7.dist-info → cartesia-2.0.0b8.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cartesia
-Version: 2.0.0b7
+Version: 2.0.0b8
 Summary: 
 Requires-Python: >=3.8,<4.0
 Classifier: Intended Audience :: Developers
@@ -47,53 +47,6 @@ Our complete API documentation can be found [on docs.cartesia.ai](https://docs.c
 pip install cartesia
 ```
 
-## Reference
-
-A full reference for this library is available [here](./reference.md).
-
-## Voices
-
-```python
-from cartesia import Cartesia
-import os
-
-client = Cartesia(api_key=os.getenv("CARTESIA_API_KEY"))
-
-# Get all available voices
-voices = client.voices.list()
-print(voices)
-
-# Get a specific voice
-voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
-print("The embedding for", voice.name, "is", voice.embedding)
-
-# Clone a voice using file data
-cloned_voice = client.voices.clone(
-    clip=open("path/to/voice.wav", "rb"),
-    name="Test cloned voice",
-    language="en",
-    mode="similarity",  # or "stability"
-    enhance=False, # use enhance=True to clean and denoise the cloning audio
-    description="Test voice description"
-)
-
-# Mix voices together
-mixed_voice = client.voices.mix(
-    voices=[
-        {"id": "voice_id_1", "weight": 0.25},
-        {"id": "voice_id_2", "weight": 0.75}
-    ]
-)
-
-# Create a new voice from embedding
-new_voice = client.voices.create(
-    name="Test Voice",
-    description="Test voice description",
-    embedding=[...],  # List[float] with 192 dimensions
-    language="en"
-)
-```
-
 ## Usage
 
 Instantiate and use the client with the following:
@@ -112,10 +65,6 @@ client.tts.bytes(
     voice={
         "mode": "id",
         "id": "694f9389-aac1-45b6-b726-9d9369183238",
-        "experimental_controls": {
-            "speed": 0.5,  # range between [-1.0, 1.0], or "slow", "fastest", etc.
-            "emotion": ["positivity", "curiosity:low"] # list of emotions with optional intensity
-        }
     },
     language="en",
     output_format={
@@ -176,7 +125,7 @@ except ApiError as e:
 
 ## Streaming
 
-The SDK supports streaming responses, as well, the response will be a generator that you can loop over.
+The SDK supports streaming responses as well, returning a generator that you can iterate over with a `for ... in ...` loop:
 
 ```python
 from cartesia import Cartesia
@@ -215,7 +164,9 @@ for chunk in chunks:
     print(f"Received chunk of size: {len(chunk.data)}")
 ```
 
-## WebSocket
+## WebSockets
+
+For the lowest latency in advanced usecases (such as streaming in an LLM-generated transcript and streaming out audio), you should use our websockets client:
 
 ```python
 from cartesia import Cartesia
@@ -223,15 +174,10 @@ from cartesia.tts import TtsRequestEmbeddingSpecifierParams, OutputFormat_RawPar
 import pyaudio
 import os
 
-client = Cartesia(
-    api_key=os.getenv("CARTESIA_API_KEY"),
-)
+client = Cartesia(api_key=os.getenv("CARTESIA_API_KEY"))
 voice_id = "a0e99841-438c-4a64-b679-ae501e7d6091"
 transcript = "Hello! Welcome to Cartesia"
 
-# You can check out our models at https://docs.cartesia.ai/getting-started/available-models
-model_id = "sonic-2"
-
 p = pyaudio.PyAudio()
 rate = 22050
 
@@ -242,14 +188,14 @@ ws = client.tts.websocket()
 
 # Generate and stream audio using the websocket
 for output in ws.send(
-    model_id=model_id,
+    model_id="sonic-2", # see: https://docs.cartesia.ai/getting-started/available-models
     transcript=transcript,
     voice={"id": voice_id},
     stream=True,
     output_format={
         "container": "raw",
         "encoding": "pcm_f32le",
-        "sample_rate": 22050
+        "sample_rate": rate
     },
 ):
     buffer = output.audio
@@ -267,6 +213,40 @@ p.terminate()
 ws.close()  # Close the websocket connection
 ```
 
+## Voices
+
+List all available Voices with `client.voices.list`, which returns an iterable that automatically handles pagination:
+
+```python
+from cartesia import Cartesia
+import os
+
+client = Cartesia(api_key=os.getenv("CARTESIA_API_KEY"))
+
+# Get all available Voices
+voices = client.voices.list()
+for voice in voices:
+    print(voice)
+```
+
+You can also get the complete metadata for a specific Voice, or make a new Voice by cloning from an audio sample:
+
+```python
+# Get a specific Voice
+voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
+print("The embedding for", voice.name, "is", voice.embedding)
+
+# Clone a Voice using file data
+cloned_voice = client.voices.clone(
+    clip=open("path/to/voice.wav", "rb"),
+    name="Test cloned voice",
+    language="en",
+    mode="similarity",  # or "stability"
+    enhance=False, # use enhance=True to clean and denoise the cloning audio
+    description="Test voice description"
+)
+```
+
 ## Requesting Timestamps
 
 ```python
@@ -290,7 +270,8 @@ async def main():
             "encoding": "pcm_f32le",
             "sample_rate": 44100
         },
-        add_timestamps=True,  # Enable word-level timestamps
+        add_timestamps=True,            # Enable word-level timestamps
+        add_phoneme_timestamps=True,    # Enable phonemized timestamps
         stream=True
     )
 
@@ -358,6 +339,26 @@ client.tts.bytes(..., request_options={
 })
 ```
 
+### Mixing voices and creating from embeddings
+
+```python
+# Mix voices together
+mixed_voice = client.voices.mix(
+    voices=[
+        {"id": "voice_id_1", "weight": 0.25},
+        {"id": "voice_id_2", "weight": 0.75}
+    ]
+)
+
+# Create a new voice from embedding
+new_voice = client.voices.create(
+    name="Test Voice",
+    description="Test voice description",
+    embedding=[...],  # List[float] with 192 dimensions
+    language="en"
+)
+```
+
 ### Custom Client
 
 You can override the `httpx` client to customize it for your use-case. Some common use-cases include support for proxies
@@ -375,6 +376,10 @@ client = Cartesia(
 )
 ```
 
+## Reference
+
+A full reference for this library is available [here](./reference.md).
+
 ## Contributing
 
 Note that most of this library is generated programmatically from

{cartesia-2.0.0b7.dist-info → cartesia-2.0.0b8.dist-info}/RECORD

@@ -1,4 +1,4 @@
-cartesia/__init__.py,sha256=r67mE_XxfYDeojBqnpfBgpAo1FnUESbX7Qm-7Vjes_Q,7965
+cartesia/__init__.py,sha256=k-YMKYUtzKObkF9Zn0TuHTC2_Z07mH6CTnZmn1my7po,8143
 cartesia/api_status/__init__.py,sha256=_dHNLdknrBjxHtU2PvLumttJM-JTQhJQqhhAQkLqt_U,168
 cartesia/api_status/client.py,sha256=GJ9Dq8iCn3hn8vCIqc6k1fCGEhSz0T0kaPGcdFnbMDY,3146
 cartesia/api_status/requests/__init__.py,sha256=ilEMzEy1JEw484CuL92bX5lHGOznc62pjiDMgiZ0tKM,130
@@ -9,7 +9,7 @@ cartesia/base_client.py,sha256=EIfMrSkJgMCgzYWJ5GN2RxsWikxcH0kMmcb3WYqfQ_g,6321
 cartesia/client.py,sha256=sPAYQLt9W2E_2F17ooocvvJImuNyLrL8xUypgf6dZeI,6238
 cartesia/core/__init__.py,sha256=-t9txgeQZL_1FDw_08GEoj4ft1Cn9Dti6X0Drsadlr0,1519
 cartesia/core/api_error.py,sha256=RE8LELok2QCjABadECTvtDp7qejA1VmINCh6TbqPwSE,426
-cartesia/core/client_wrapper.py,sha256=tTxN1WEjVJuMSKTZ4kVKQykuql_lcQuiUfDU89z-f0A,1856
+cartesia/core/client_wrapper.py,sha256=BEIOireABuSTdCAcsHeQKtZ1D3sIi-CVQv5YFHmfi3Y,1856
 cartesia/core/datetime_utils.py,sha256=nBys2IsYrhPdszxGKCNRPSOCwa-5DWOHG95FB8G9PKo,1047
 cartesia/core/file.py,sha256=d4NNbX8XvXP32z8KpK2Xovv33nFfruIrpz0QWxlgpZk,2663
 cartesia/core/http_client.py,sha256=KL5RGa0y4n8nX0-07WRg4ZQUTq30sc-XJbWcP5vjBDg,19552
@@ -42,8 +42,8 @@ cartesia/infill/__init__.py,sha256=FTtvy8EDg9nNNg9WCatVgKTRYV8-_v1roeGPAKoa_pw,6
 cartesia/infill/client.py,sha256=PWE5Ak-wsaBM_8g52oDl9PYx76PkW6f900mnxvZf4Bk,12571
 cartesia/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 cartesia/tts/__init__.py,sha256=G0wcYlPrr7hmu5DQgCG7bDTQq36fpP3iBM5164Z0-Js,4701
-cartesia/tts/_async_websocket.py,sha256=mPykrS40FJee58T8NtGUnQ7AurQy04Qz6ICzjCnr7Fg,18383
-cartesia/tts/_websocket.py,sha256=Gzd2GvTPUKn59u7quVHn53cGe44H_fCv1jr-opSHRZk,18689
+cartesia/tts/_async_websocket.py,sha256=U7ySTJqb3V0RDSKPcFfzpBa0pqui05k5BTqiIpSBth0,18652
+cartesia/tts/_websocket.py,sha256=roMJ7oDSjr5U5sTHM8EcGu-EtzbIVUH4HmOY1yI2JL4,19118
 cartesia/tts/client.py,sha256=KMhDaW0gG_uwkSq1EzoC-bCx1G0TLB4K4Gm57L4xDSs,14832
 cartesia/tts/requests/__init__.py,sha256=0rcfMLHNbUhkRI1xS09UE4p-WT1BCqrcblFtPxcATOI,3261
 cartesia/tts/requests/cancel_context_request.py,sha256=Wl8g-o5vwl9ENm-H1wsLx441FkIR_4Wt5UYtuWce2Yw,431
@@ -120,35 +120,38 @@ cartesia/voice_changer/requests/streaming_response.py,sha256=cV6L9mMY0w2JpJ0xKoF
 cartesia/voice_changer/types/__init__.py,sha256=qAiHsdRpnFeS0lBkYp_NRrhSJiRXCg5-uFibqDWzYVU,430
 cartesia/voice_changer/types/output_format_container.py,sha256=RqLDELdgeOjYqNTJX1Le62qjiFiJGxf0cYnol88-LLM,166
 cartesia/voice_changer/types/streaming_response.py,sha256=rQ4ZehtOHsCBKijyULz_ahGQYNj1yus6AM6u2wgcBsI,1963
-cartesia/voices/__init__.py,sha256=ipS0rBobAU31yoJEbZ-2LcENhmmpzjxfzc_h5v3R0zk,1713
-cartesia/voices/client.py,sha256=nOmRRJevMyBtmuTNa6aDFWpQXu1GFkjNdfzFrMMwl5k,37160
+cartesia/voices/__init__.py,sha256=2D58Bir45LvcvP08QMnPlFE8DD8BONTjPLkIDdKs7vg,1891
+cartesia/voices/client.py,sha256=8zQZAtaCAJi79puMxVhzR5OWCDjows53k4oTvSgcdJM,38867
 cartesia/voices/requests/__init__.py,sha256=XiBJbSYeQCgFMtwywKvQ0Nmp7Zf_0WskzRhgr9c8h38,1072
-cartesia/voices/requests/create_voice_request.py,sha256=HvxxWBwR5RMMMmxEU5Tj5jsDSXnlT0cS-C6AGlMPlr0,509
+cartesia/voices/requests/create_voice_request.py,sha256=r6dKb9ga0ZsAi_6PXuE43u2lLgfQg2DIYjk2Neng7pI,617
 cartesia/voices/requests/embedding_response.py,sha256=PGZkBD8UBcv2MYQbBXyD4T6lzaE9oSGGwXx-MoXCp0M,228
 cartesia/voices/requests/embedding_specifier.py,sha256=PAHdGsVmLLeJC2b1fWHWI_OlhogO1WnJdzoX9pj5N8c,282
 cartesia/voices/requests/get_voices_response.py,sha256=g-ZCaCaLOlZSitcKVhdCtfdKQQz8N3W6E7_wZUNOi5M,747
 cartesia/voices/requests/id_specifier.py,sha256=UTtoXBEEYaGvg-Dn2QxUDACNB3Vm1O1XbrPtBA3rGzU,252
-cartesia/voices/requests/localize_dialect.py,sha256=9mmLHOFbBvWZoU2PyjXozG6hoDpE0uueymXHi0k_VtE,209
-cartesia/voices/requests/localize_voice_request.py,sha256=AkY4cvx31MF3_gkqMpUzibGIOh9cNF5cOCf3Yqnm7Vc,549
+cartesia/voices/requests/localize_dialect.py,sha256=OHAInU6IP0LBzIY3VYSiU9bRLjXfr1pGXunsLgv1QHs,497
+cartesia/voices/requests/localize_voice_request.py,sha256=oh828eqYkiticD_lerc8WemN3bW13mLZpfRDiKbG75g,703
 cartesia/voices/requests/mix_voice_specifier.py,sha256=YjOJ2Qt3nqMQzHsYbF1DnZgmZS9zZepLXpji6V9mfgs,266
 cartesia/voices/requests/mix_voices_request.py,sha256=6JCzFmWKIS1_t-uSoO1m-FQbLWB1zaykTcGV-1s-RqM,275
 cartesia/voices/requests/update_voice_request.py,sha256=XxJ6TKO4M2s1kXQAZRj8uA4okIABvmWiFhAHJv4BS0Q,282
 cartesia/voices/requests/voice.py,sha256=M-4lf4W57fx84_JFOy55b9mWcqO4LfzpY-G_Ekv-2Bo,1031
 cartesia/voices/requests/voice_metadata.py,sha256=S0jPQtBpEb2WSnYDLQTS7pcbNJpc0d01uWravHaqzso,697
-cartesia/voices/types/__init__.py,sha256=fsPgm1Ma1E_iBIKUMseIie9QrcGD-p31_KeMvPMb_KA,1503
+cartesia/voices/types/__init__.py,sha256=yjxMWjoBpwAZ5UJ2iRSC_kKgZvGmqVd09kQxgcTnMac,1782
 cartesia/voices/types/base_voice_id.py,sha256=nWRC0rvLpjeMpRbLSmUTPziWo1ZrbPxw22l4gEBWp8Q,118
 cartesia/voices/types/clone_mode.py,sha256=3sR6wdxym4xDVsoHppp3-V9mpDwP9F9fDfMUQKG24xw,160
-cartesia/voices/types/create_voice_request.py,sha256=8vfKu6cD_VYFb3GN5gVpxlRUIZALYE-449NbDSnXaDg,911
+cartesia/voices/types/create_voice_request.py,sha256=_q0d8QojmQrpU-Puzd_YvWmiC7cBp_lrbKmTLuknYqQ,1005
 cartesia/voices/types/embedding_response.py,sha256=B7MJ79HIAnxtiP6OT0tt27KBDYTZ3VU0MLuQfb5qVOg,624
 cartesia/voices/types/embedding_specifier.py,sha256=cf6JfVnISyrvjWup3oAg-RFdMVRxytem6HLwZgKl3gA,671
 cartesia/voices/types/gender.py,sha256=OrbTO__3HVNculvkcb5Pz-Yoa-Xv8N_rNMrFoy2DoaA,148
 cartesia/voices/types/gender_presentation.py,sha256=rM8pSurYCSH0AGgLsVpVAPp7uz7TQMM1nPa7-Vus7gw,185
 cartesia/voices/types/get_voices_response.py,sha256=c6KMkmJepTUmT7I6tAVOGrPst2kkXxDCXLIf1AnR9NE,1136
 cartesia/voices/types/id_specifier.py,sha256=yAY-uc9hRJkHXdsSfRZWkE8ga2Sb-KVipOTSXa8Wmp0,634
-cartesia/voices/types/localize_dialect.py,sha256=tRckNEq4EsdYPondF1rrjOrYRZUSL6WW_3627cFwG1I,196
+cartesia/voices/types/localize_dialect.py,sha256=6JpJKeQvtDjCT2n-5yaGOe3D-4nYqUoYrvcCSE2Zxik,463
 cartesia/voices/types/localize_english_dialect.py,sha256=0PjZNjQv5ll2wWZxGveQIYCUGLtGDVELK9FBWFe7SNc,176
+cartesia/voices/types/localize_french_dialect.py,sha256=aMhqLi_5goAaSGZguZIFOwQ9Yqh5ApL6gS3cDI315lQ,157
+cartesia/voices/types/localize_portuguese_dialect.py,sha256=6dcThK1qWyS3c-W--3Zz7HK5ixS0qslEWrVQmKSrl9E,161
+cartesia/voices/types/localize_spanish_dialect.py,sha256=h-H52vk0MBOvJqlzPVPgajfQU6oxpTzHoQAKmSDyaC4,158
 cartesia/voices/types/localize_target_language.py,sha256=ttngtFVpMvuWAKQztJu_pCaf7V62DzmNq9zthPCb2LI,242
-cartesia/voices/types/localize_voice_request.py,sha256=roZkcA7LiYs_L1R9FgTCTIgmHv9TUfXZMgLEnrajJ3I,887
+cartesia/voices/types/localize_voice_request.py,sha256=gvjg292kMgji0L9TNO3VqDS0pHO1vGJUcf0l_vEW_5Y,1098
 cartesia/voices/types/mix_voice_specifier.py,sha256=B0FE6UREGk1TxlN0GOPwyCuqJbMkWVUs0EFqiJuQfZ8,236
 cartesia/voices/types/mix_voices_request.py,sha256=R_8bmUmE1br4wmfH1Qu6EnL9uC-V1z5BV3_B7u51EOw,641
 cartesia/voices/types/update_voice_request.py,sha256=_CEH8nuSZn2qZa9xZlANZXOhJd49XLel3dRy2dfOvr8,716
@@ -157,6 +160,6 @@ cartesia/voices/types/voice_expand_options.py,sha256=e4FroWdlxEE-LXQfT1RWlGHtswl
 cartesia/voices/types/voice_id.py,sha256=GDoXcRVeIm-V21R4suxG2zqLD3DLYkXE9kgizadzFKo,79
 cartesia/voices/types/voice_metadata.py,sha256=4KNGjXMUKm3niv-NvKIFVGtiilpH13heuzKcZYNQxk4,1181
 cartesia/voices/types/weight.py,sha256=XqDU7_JItNUb5QykIDqTbELlRYQdbt2SviRgW0w2LKo,80
-cartesia-2.0.0b7.dist-info/METADATA,sha256=8sWG16O3-gGLZWHd8FrRQRru7cKLH4RiHCr0uEWzqd0,10895
-cartesia-2.0.0b7.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
-cartesia-2.0.0b7.dist-info/RECORD,,
+cartesia-2.0.0b8.dist-info/METADATA,sha256=ynQsxGb1v5ZHMnXkeqYceRFrC-bxwuRaopOPyuBbCsk,11208
+cartesia-2.0.0b8.dist-info/WHEEL,sha256=Zb28QaM1gQi8f4VCBhsUklF61CTlNYfs9YAZn-TOGFk,88
+cartesia-2.0.0b8.dist-info/RECORD,,