cartesia 1.0.8__tar.gz → 1.0.10__tar.gz

{cartesia-1.0.8 → cartesia-1.0.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cartesia
-Version: 1.0.8
+Version: 1.0.10
 Summary: The official Python library for the Cartesia API.
 Home-page: 
 Author: Cartesia, Inc.

cartesia-1.0.10/cartesia/__init__.py

@@ -0,0 +1,3 @@
+from cartesia.client import AsyncCartesia, Cartesia
+
+__all__ = ["Cartesia", "AsyncCartesia"]

{cartesia-1.0.8 → cartesia-1.0.10}/cartesia/_types.py

@@ -1,4 +1,5 @@
-from typing import List, TypedDict
+from typing import List, TypedDict, Union
+
 from cartesia.utils.deprecated import deprecated
 
 
@@ -85,7 +86,7 @@ class VoiceControls(TypedDict):
         This is an experimental class and is subject to rapid change in future versions.
     """
 
-    speed: str = ""
+    speed: Union[str, float] = ""
     emotion: List[str] = []
 
 

{cartesia-1.0.8 → cartesia-1.0.10}/cartesia/client.py

@@ -1,46 +1,48 @@
 import asyncio
 import base64
-from collections import defaultdict
 import json
+import logging
 import os
 import uuid
+from collections import defaultdict
 from types import TracebackType
 from typing import (
     Any,
     AsyncGenerator,
-    Iterator,
+    Callable,
     Dict,
     Generator,
+    Iterator,
     List,
     Optional,
+    Set,
     Tuple,
     Union,
-    Callable,
-    Set,
 )
 
 import aiohttp
 import httpx
-import logging
 import requests
+
 try:
     from websockets.sync.client import connect
+
     IS_WEBSOCKET_SYNC_AVAILABLE = True
 except ImportError:
     IS_WEBSOCKET_SYNC_AVAILABLE = False
 
 from iterators import TimeoutIterator
+from websockets.sync.client import connect
 
-from cartesia.utils.retry import retry_on_connection_error, retry_on_connection_error_async
 from cartesia._types import (
+    DeprecatedOutputFormatMapping,
     EventType,
     OutputFormat,
     OutputFormatMapping,
-    DeprecatedOutputFormatMapping,
     VoiceControls,
     VoiceMetadata,
 )
-
+from cartesia.utils.retry import retry_on_connection_error, retry_on_connection_error_async
 
 DEFAULT_MODEL_ID = "sonic-english"  # latest default model
 MULTILINGUAL_MODEL_ID = "sonic-multilingual"  # latest multilingual model
@@ -212,21 +214,22 @@ class Voices(Resource):
 
         return response.json()
 
-    def clone(self, filepath: Optional[str] = None, link: Optional[str] = None) -> List[float]:
+    def clone(self, filepath: Optional[str] = None, enhance: str = True) -> List[float]:
         """Clone a voice from a clip.
 
         Args:
             filepath: The path to the clip file.
+            enhance: Whether to enhance the clip before cloning the voice (highly recommended). Defaults to True.
 
         Returns:
             The embedding of the cloned voice as a list of floats.
         """
-        # TODO: Python has a bytes object, use that instead of a filepath
         if not filepath:
             raise ValueError("Filepath must be specified.")
         url = f"{self._http_url()}/voices/clone/clip"
         with open(filepath, "rb") as file:
             files = {"clip": file}
+            files["enhance"] = str(enhance).lower()
             headers = self.headers.copy()
             headers.pop("Content-Type", None)
             response = httpx.post(url, headers=headers, files=files, timeout=self.timeout)

cartesia-1.0.10/cartesia/version.py

@@ -0,0 +1 @@
+__version__ = "1.0.10"

{cartesia-1.0.8 → cartesia-1.0.10}/cartesia.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: cartesia
-Version: 1.0.8
+Version: 1.0.10
 Summary: The official Python library for the Cartesia API.
 Home-page: 
 Author: Cartesia, Inc.

{cartesia-1.0.8 → cartesia-1.0.10}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.ruff]
 # Add more rule codes as needed
-extend-select = [
+lint.extend-select = [
     "D",    # pydocstyle - to replace docformatter
 ]
 
@@ -43,14 +43,14 @@ line-length = 100
 # Enable the count of violations
 output-format = "full"
 
-[tool.ruff.isort]
-force-wrap-aliases = true
-combine-as-imports = true
-force-sort-within-sections = true
-known-first-party = []
-known-third-party = []
-known-local-folder = []
-lines-after-imports = 2
-
-[tool.ruff.pydocstyle]
+[tool.ruff.lint.pydocstyle]
 convention = "google"
+
+[tool.isort]
+profile = "black"
+multi_line_output = 3
+include_trailing_comma = true
+force_grid_wrap = 0
+use_parentheses = true
+ensure_newline_before_comments = true
+line_length = 100

{cartesia-1.0.8 → cartesia-1.0.10}/tests/test_deprecated.py

@@ -1,8 +1,7 @@
 from packaging.version import Version
 
-import cartesia as Cartesia
-from cartesia.utils.deprecated import _DEPRECATED_FUNCTION_STATS
 import cartesia.version as version
+from cartesia.utils.deprecated import _DEPRECATED_FUNCTION_STATS
 
 
 def test_deprecated_to_remove_by_version():

{cartesia-1.0.8 → cartesia-1.0.10}/tests/test_tts.py

@@ -5,17 +5,19 @@ different results. Therefore, we cannot test for complete correctness but rather
 general correctness.
 """
 
+import asyncio
 import logging
 import os
 import sys
-from cartesia import AsyncCartesia, Cartesia
-from cartesia.client import DEFAULT_MODEL_ID, MULTILINGUAL_MODEL_ID
-from cartesia._types import VoiceControls, VoiceMetadata
+import uuid
 from typing import AsyncGenerator, Generator, List
+
 import numpy as np
 import pytest
-import uuid
-import asyncio
+
+from cartesia import AsyncCartesia, Cartesia
+from cartesia._types import VoiceControls, VoiceMetadata
+from cartesia.client import DEFAULT_MODEL_ID, MULTILINGUAL_MODEL_ID
 
 THISDIR = os.path.dirname(__file__)
 sys.path.insert(0, os.path.dirname(THISDIR))
@@ -24,6 +26,7 @@ RESOURCES_DIR = os.path.join(THISDIR, "resources")
 SAMPLE_VOICE = "Newsman"
 SAMPLE_VOICE_ID = "d46abd1d-2d02-43e8-819f-51fb652c1c61"
 EXPERIMENTAL_VOICE_CONTROLS = {"emotion": ["anger:high", "positivity:low"], "speed": "fastest"}
+EXPERIMENTAL_VOICE_CONTROLS_2 = {"speed": 0.4}
 
 logger = logging.getLogger(__name__)
 
@@ -84,6 +87,12 @@ def test_clone_voice_with_file(client: Cartesia):
     output = client.voices.clone(filepath=os.path.join(RESOURCES_DIR, "sample-speech-4s.wav"))
     assert isinstance(output, list)
 
+@pytest.mark.parametrize("enhance", [True, False])
+def test_clone_voice_with_file_enhance(client: Cartesia, enhance: bool):
+    logger.info("Testing voices.clone with file")
+    output = client.voices.clone(filepath=os.path.join(RESOURCES_DIR, "sample-speech-4s.wav"), enhance=enhance)
+    assert isinstance(output, list)
+
 def test_create_voice(client: Cartesia):
     logger.info("Testing voices.create")
     embedding = np.ones(192).tolist()
@@ -95,7 +104,7 @@ def test_create_voice(client: Cartesia):
     assert voice in voices
 
 @pytest.mark.parametrize("stream", [True, False])
-@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS])
+@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS, EXPERIMENTAL_VOICE_CONTROLS_2])
 def test_sse_send(resources: _Resources, stream: bool, _experimental_voice_controls: VoiceControls):
     logger.info("Testing SSE send")
     client = resources.client
@@ -132,7 +141,7 @@ def test_sse_send_with_model_id(resources: _Resources, stream: bool):
         assert isinstance(out["audio"], bytes)
         
 @pytest.mark.parametrize("stream", [True, False])
-@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS])
+@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS, EXPERIMENTAL_VOICE_CONTROLS_2])
 def test_websocket_send(resources: _Resources, stream: bool, _experimental_voice_controls: VoiceControls):
     logger.info("Testing WebSocket send")
     client = resources.client
@@ -182,7 +191,7 @@ def test_websocket_send_timestamps(resources: _Resources, stream: bool):
     ws.close()
 
         
-@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS])
+@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS, EXPERIMENTAL_VOICE_CONTROLS_2])
 def test_sse_send_context_manager(resources: _Resources, _experimental_voice_controls: VoiceControls):
     logger.info("Testing SSE send context manager")
     transcript = "Hello, world! I'\''m generating audio on Cartesia."
@@ -248,7 +257,7 @@ def test_websocket_send_context_manage_err(resources: _Resources):
         pass
 
 @pytest.mark.asyncio
-@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS])
+@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS, EXPERIMENTAL_VOICE_CONTROLS_2])
 async def test_async_sse_send( resources: _Resources, _experimental_voice_controls: VoiceControls):
     logger.info("Testing async SSE send")
     transcript = "Hello, world! I'\''m generating audio on Cartesia."
@@ -269,7 +278,7 @@ async def test_async_sse_send( resources: _Resources, _experimental_voice_contro
         await async_client.close()
         
 @pytest.mark.asyncio
-@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS])
+@pytest.mark.parametrize("_experimental_voice_controls", [None, EXPERIMENTAL_VOICE_CONTROLS, EXPERIMENTAL_VOICE_CONTROLS_2])
 async def test_async_websocket_send(resources: _Resources,  _experimental_voice_controls: VoiceControls):
     logger.info("Testing async WebSocket send")
     transcript = "Hello, world! I'\''m generating audio on Cartesia."

cartesia-1.0.8/cartesia/__init__.py

@@ -1,3 +0,0 @@
-from cartesia.client import Cartesia, AsyncCartesia
-
-__all__ = ["Cartesia", "AsyncCartesia"]

cartesia-1.0.8/cartesia/version.py

@@ -1 +0,0 @@
-__version__ = "1.0.8"

{cartesia-1.0.8 → cartesia-1.0.10}/cartesia/utils/retry.py

@@ -1,9 +1,9 @@
-import time
-
-from aiohttp.client_exceptions import ServerDisconnectedError
 import asyncio
+import time
 from functools import wraps
 from http.client import RemoteDisconnected
+
+from aiohttp.client_exceptions import ServerDisconnectedError
 from httpx import TimeoutException
 from requests.exceptions import ConnectionError