cartesia 1.3.1__tar.gz → 2.0.0a0__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- cartesia-2.0.0a0/PKG-INFO +306 -0
- cartesia-2.0.0a0/README.md +274 -0
- cartesia-2.0.0a0/pyproject.toml +70 -0
- cartesia-2.0.0a0/src/cartesia/__init__.py +289 -0
- cartesia-2.0.0a0/src/cartesia/api_status/__init__.py +6 -0
- cartesia-2.0.0a0/src/cartesia/api_status/client.py +104 -0
- cartesia-2.0.0a0/src/cartesia/api_status/requests/__init__.py +5 -0
- cartesia-2.0.0a0/src/cartesia/api_status/requests/api_info.py +8 -0
- cartesia-2.0.0a0/src/cartesia/api_status/types/__init__.py +5 -0
- cartesia-2.0.0a0/src/cartesia/api_status/types/api_info.py +20 -0
- cartesia-2.0.0a0/src/cartesia/base_client.py +160 -0
- cartesia-2.0.0a0/src/cartesia/client.py +192 -0
- cartesia-2.0.0a0/src/cartesia/core/__init__.py +47 -0
- cartesia-2.0.0a0/src/cartesia/core/api_error.py +15 -0
- cartesia-2.0.0a0/src/cartesia/core/client_wrapper.py +55 -0
- cartesia-2.0.0a0/src/cartesia/core/datetime_utils.py +28 -0
- cartesia-2.0.0a0/src/cartesia/core/file.py +67 -0
- cartesia-2.0.0a0/src/cartesia/core/http_client.py +499 -0
- cartesia-2.0.0a0/src/cartesia/core/jsonable_encoder.py +101 -0
- cartesia-2.0.0a0/src/cartesia/core/pydantic_utilities.py +296 -0
- cartesia-2.0.0a0/src/cartesia/core/query_encoder.py +58 -0
- cartesia-2.0.0a0/src/cartesia/core/remove_none_from_dict.py +11 -0
- cartesia-2.0.0a0/src/cartesia/core/request_options.py +35 -0
- cartesia-2.0.0a0/src/cartesia/core/serialization.py +272 -0
- cartesia-2.0.0a0/src/cartesia/datasets/__init__.py +24 -0
- cartesia-2.0.0a0/src/cartesia/datasets/client.py +422 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/__init__.py +15 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/create_dataset_request.py +7 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/dataset.py +9 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/dataset_file.py +9 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/paginated_dataset_files.py +10 -0
- cartesia-2.0.0a0/src/cartesia/datasets/requests/paginated_datasets.py +10 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/__init__.py +17 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/create_dataset_request.py +19 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/dataset.py +21 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/dataset_file.py +21 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/file_purpose.py +5 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/paginated_dataset_files.py +21 -0
- cartesia-2.0.0a0/src/cartesia/datasets/types/paginated_datasets.py +21 -0
- cartesia-2.0.0a0/src/cartesia/embedding/__init__.py +5 -0
- cartesia-2.0.0a0/src/cartesia/embedding/types/__init__.py +5 -0
- cartesia-2.0.0a0/src/cartesia/embedding/types/embedding.py +201 -0
- cartesia-2.0.0a0/src/cartesia/environment.py +7 -0
- cartesia-2.0.0a0/src/cartesia/infill/__init__.py +2 -0
- cartesia-2.0.0a0/src/cartesia/infill/client.py +294 -0
- cartesia-2.0.0a0/src/cartesia/tts/__init__.py +167 -0
- {cartesia-1.3.1/cartesia → cartesia-2.0.0a0/src/cartesia/tts}/_async_websocket.py +159 -84
- cartesia-2.0.0a0/src/cartesia/tts/_websocket.py +430 -0
- cartesia-2.0.0a0/src/cartesia/tts/client.py +407 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/__init__.py +76 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/cancel_context_request.py +17 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/controls.py +11 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/generation_request.py +53 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/mp_3_output_format.py +11 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/output_format.py +30 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/phoneme_timestamps.py +10 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/raw_output_format.py +11 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/speed.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/tts_request.py +24 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/tts_request_embedding_specifier.py +16 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/tts_request_id_specifier.py +16 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/tts_request_voice_specifier.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/wav_output_format.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_base_response.py +11 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_chunk_response.py +8 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_done_response.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_error_response.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_flush_done_response.py +9 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_phoneme_timestamps_response.py +9 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_raw_output_format.py +11 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_request.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_response.py +69 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_stream_options.py +8 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_timestamps_response.py +9 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_tts_output.py +18 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/web_socket_tts_request.py +24 -0
- cartesia-2.0.0a0/src/cartesia/tts/requests/word_timestamps.py +10 -0
- cartesia-2.0.0a0/src/cartesia/tts/socket_client.py +302 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/__init__.py +90 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/cancel_context_request.py +28 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/context_id.py +3 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/controls.py +22 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/emotion.py +29 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/flush_id.py +3 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/generation_request.py +66 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/mp_3_output_format.py +23 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/natural_specifier.py +5 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/numerical_specifier.py +3 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/output_format.py +58 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/phoneme_timestamps.py +21 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/raw_encoding.py +5 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/raw_output_format.py +22 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/speed.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/supported_language.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/tts_request.py +35 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/tts_request_embedding_specifier.py +27 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/tts_request_id_specifier.py +27 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/tts_request_voice_specifier.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/wav_output_format.py +17 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_base_response.py +22 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_chunk_response.py +20 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_done_response.py +17 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_error_response.py +19 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_flush_done_response.py +21 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_phoneme_timestamps_response.py +20 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_raw_output_format.py +22 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_request.py +7 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_response.py +124 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_stream_options.py +19 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_timestamps_response.py +20 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_tts_output.py +27 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/web_socket_tts_request.py +36 -0
- cartesia-2.0.0a0/src/cartesia/tts/types/word_timestamps.py +21 -0
- cartesia-2.0.0a0/src/cartesia/tts/utils/tts.py +64 -0
- cartesia-2.0.0a0/src/cartesia/tts/utils/types.py +70 -0
- cartesia-2.0.0a0/src/cartesia/version.py +3 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/__init__.py +27 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/client.py +395 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/requests/__init__.py +15 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/requests/streaming_response.py +36 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/types/__init__.py +17 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/types/output_format_container.py +5 -0
- cartesia-2.0.0a0/src/cartesia/voice_changer/types/streaming_response.py +62 -0
- cartesia-2.0.0a0/src/cartesia/voices/__init__.py +67 -0
- cartesia-2.0.0a0/src/cartesia/voices/client.py +1812 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/__init__.py +27 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/create_voice_request.py +21 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/embedding_response.py +8 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/embedding_specifier.py +10 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/id_specifier.py +10 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/localize_dialect.py +6 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/localize_voice_request.py +15 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/mix_voice_specifier.py +7 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/mix_voices_request.py +9 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/update_voice_request.py +15 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/voice.py +39 -0
- cartesia-2.0.0a0/src/cartesia/voices/requests/voice_metadata.py +36 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/__init__.py +41 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/base_voice_id.py +5 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/clone_mode.py +5 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/create_voice_request.py +32 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/embedding_response.py +20 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/embedding_specifier.py +22 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/gender.py +5 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/id_specifier.py +22 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/localize_dialect.py +6 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/localize_english_dialect.py +5 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/localize_target_language.py +7 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/localize_voice_request.py +26 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/mix_voice_specifier.py +7 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/mix_voices_request.py +20 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/update_voice_request.py +27 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/voice.py +50 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/voice_id.py +3 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/voice_metadata.py +48 -0
- cartesia-2.0.0a0/src/cartesia/voices/types/weight.py +3 -0
- cartesia-1.3.1/.github/workflows/ci.yaml +0 -64
- cartesia-1.3.1/.github/workflows/publish.yaml +0 -58
- cartesia-1.3.1/.gitignore +0 -60
- cartesia-1.3.1/LICENSE.md +0 -21
- cartesia-1.3.1/Makefile +0 -13
- cartesia-1.3.1/PKG-INFO +0 -661
- cartesia-1.3.1/README.md +0 -649
- cartesia-1.3.1/bumpversion.py +0 -41
- cartesia-1.3.1/cartesia/__init__.py +0 -4
- cartesia-1.3.1/cartesia/_async_sse.py +0 -95
- cartesia-1.3.1/cartesia/_logger.py +0 -3
- cartesia-1.3.1/cartesia/_sse.py +0 -143
- cartesia-1.3.1/cartesia/_types.py +0 -70
- cartesia-1.3.1/cartesia/_websocket.py +0 -358
- cartesia-1.3.1/cartesia/async_client.py +0 -82
- cartesia-1.3.1/cartesia/async_tts.py +0 -63
- cartesia-1.3.1/cartesia/client.py +0 -69
- cartesia-1.3.1/cartesia/resource.py +0 -44
- cartesia-1.3.1/cartesia/tts.py +0 -137
- cartesia-1.3.1/cartesia/utils/deprecated.py +0 -55
- cartesia-1.3.1/cartesia/utils/retry.py +0 -87
- cartesia-1.3.1/cartesia/utils/tts.py +0 -78
- cartesia-1.3.1/cartesia/version.py +0 -1
- cartesia-1.3.1/cartesia/voices.py +0 -208
- cartesia-1.3.1/pyproject.toml +0 -88
- cartesia-1.3.1/tests/__init__.py +0 -0
- cartesia-1.3.1/tests/resources/sample-speech-4s.wav +0 -0
- cartesia-1.3.1/tests/test_deprecated.py +0 -20
- cartesia-1.3.1/tests/test_tts.py +0 -1170
- cartesia-1.3.1/uv.lock +0 -1454
- /cartesia-1.3.1/cartesia/utils/__init__.py → /cartesia-2.0.0a0/src/cartesia/py.typed +0 -0
- /cartesia-1.3.1/cartesia/_constants.py → /cartesia-2.0.0a0/src/cartesia/tts/utils/constants.py +0 -0
@@ -0,0 +1,306 @@
|
|
1
|
+
Metadata-Version: 2.1
|
2
|
+
Name: cartesia
|
3
|
+
Version: 2.0.0a0
|
4
|
+
Summary:
|
5
|
+
Requires-Python: >=3.8,<4.0
|
6
|
+
Classifier: Intended Audience :: Developers
|
7
|
+
Classifier: Operating System :: MacOS
|
8
|
+
Classifier: Operating System :: Microsoft :: Windows
|
9
|
+
Classifier: Operating System :: OS Independent
|
10
|
+
Classifier: Operating System :: POSIX
|
11
|
+
Classifier: Operating System :: POSIX :: Linux
|
12
|
+
Classifier: Programming Language :: Python
|
13
|
+
Classifier: Programming Language :: Python :: 3
|
14
|
+
Classifier: Programming Language :: Python :: 3.8
|
15
|
+
Classifier: Programming Language :: Python :: 3.9
|
16
|
+
Classifier: Programming Language :: Python :: 3.10
|
17
|
+
Classifier: Programming Language :: Python :: 3.11
|
18
|
+
Classifier: Programming Language :: Python :: 3.12
|
19
|
+
Classifier: Topic :: Software Development :: Libraries :: Python Modules
|
20
|
+
Classifier: Typing :: Typed
|
21
|
+
Requires-Dist: aiohttp (>=3.10.10)
|
22
|
+
Requires-Dist: httpx (>=0.21.2)
|
23
|
+
Requires-Dist: httpx-sse (==0.4.0)
|
24
|
+
Requires-Dist: iterators (>=0.2.0)
|
25
|
+
Requires-Dist: pydantic (>=1.9.2)
|
26
|
+
Requires-Dist: pydantic-core (>=2.18.2,<3.0.0)
|
27
|
+
Requires-Dist: pydub (>=0.25.1)
|
28
|
+
Requires-Dist: typing_extensions (>=4.0.0)
|
29
|
+
Requires-Dist: websockets (>=10.4)
|
30
|
+
Description-Content-Type: text/markdown
|
31
|
+
|
32
|
+
# Cartesia Python Library
|
33
|
+
|
34
|
+
[](https://buildwithfern.com?utm_source=github&utm_medium=github&utm_campaign=readme&utm_source=https%3A%2F%2Fgithub.com%2Fcartesia-ai%2Fcartesia-python)
|
35
|
+
[](https://pypi.python.org/pypi/cartesia)
|
36
|
+
|
37
|
+
The Cartesia Python library provides convenient access to the Cartesia API from Python.
|
38
|
+
|
39
|
+
## Documentation
|
40
|
+
|
41
|
+
Our complete API documentation can be found [on docs.cartesia.ai](https://docs.cartesia.ai).
|
42
|
+
|
43
|
+
## Installation
|
44
|
+
|
45
|
+
```sh
|
46
|
+
pip install cartesia
|
47
|
+
```
|
48
|
+
|
49
|
+
## Reference
|
50
|
+
|
51
|
+
A full reference for this library is available [here](./reference.md).
|
52
|
+
|
53
|
+
## Voices
|
54
|
+
|
55
|
+
```python
|
56
|
+
from cartesia import Cartesia
|
57
|
+
import os
|
58
|
+
|
59
|
+
client = Cartesia(api_key=os.environ.get("CARTESIA_API_KEY"))
|
60
|
+
|
61
|
+
# Get all available voices
|
62
|
+
voices = client.voices.list()
|
63
|
+
print(voices)
|
64
|
+
|
65
|
+
# Get a specific voice
|
66
|
+
voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
|
67
|
+
print("The embedding for", voice["name"], "is", voice["embedding"])
|
68
|
+
|
69
|
+
# Clone a voice using filepath
|
70
|
+
cloned_voice_embedding = client.voices.clone(filepath="path/to/voice")
|
71
|
+
|
72
|
+
# Mix voices together
|
73
|
+
mixed_voice_embedding = client.voices.mix(
|
74
|
+
[{ "id": "voice_id_1", "weight": 0.5 }, { "id": "voice_id_2", "weight": 0.25 }, { "id": "voice_id_3", "weight": 0.25 }]
|
75
|
+
)
|
76
|
+
|
77
|
+
# Create a new voice
|
78
|
+
new_voice = client.voices.create(
|
79
|
+
name="New Voice",
|
80
|
+
description="A clone of my own voice",
|
81
|
+
embedding=cloned_voice_embedding,
|
82
|
+
)
|
83
|
+
```
|
84
|
+
|
85
|
+
## Usage
|
86
|
+
|
87
|
+
Instantiate and use the client with the following:
|
88
|
+
|
89
|
+
```python
|
90
|
+
from cartesia import Cartesia
|
91
|
+
from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
|
92
|
+
|
93
|
+
client = Cartesia(
|
94
|
+
api_key="YOUR_API_KEY",
|
95
|
+
)
|
96
|
+
client.tts.bytes(
|
97
|
+
model_id="sonic-english",
|
98
|
+
transcript="Hello, world!",
|
99
|
+
voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
|
100
|
+
),
|
101
|
+
language="en",
|
102
|
+
output_format={
|
103
|
+
"container": "raw",
|
104
|
+
"sample_rate": 44100,
|
105
|
+
"encoding": "pcm_f32le",
|
106
|
+
},
|
107
|
+
)
|
108
|
+
```
|
109
|
+
|
110
|
+
## Async Client
|
111
|
+
|
112
|
+
The SDK also exports an `async` client so that you can make non-blocking calls to our API.
|
113
|
+
|
114
|
+
```python
|
115
|
+
import asyncio
|
116
|
+
|
117
|
+
from cartesia import AsyncCartesia
|
118
|
+
from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
|
119
|
+
|
120
|
+
client = AsyncCartesia(
|
121
|
+
api_key="YOUR_API_KEY",
|
122
|
+
)
|
123
|
+
|
124
|
+
|
125
|
+
async def main() -> None:
|
126
|
+
await client.tts.bytes(
|
127
|
+
model_id="sonic-english",
|
128
|
+
transcript="Hello, world!",
|
129
|
+
voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
|
130
|
+
language="en",
|
131
|
+
output_format={
|
132
|
+
"container": "raw",
|
133
|
+
"sample_rate": 44100,
|
134
|
+
"encoding": "pcm_f32le",
|
135
|
+
},
|
136
|
+
)
|
137
|
+
|
138
|
+
|
139
|
+
asyncio.run(main())
|
140
|
+
```
|
141
|
+
|
142
|
+
## Exception Handling
|
143
|
+
|
144
|
+
When the API returns a non-success status code (4xx or 5xx response), a subclass of the following error
|
145
|
+
will be thrown.
|
146
|
+
|
147
|
+
```python
|
148
|
+
from cartesia.core.api_error import ApiError
|
149
|
+
|
150
|
+
try:
|
151
|
+
client.tts.bytes(...)
|
152
|
+
except ApiError as e:
|
153
|
+
print(e.status_code)
|
154
|
+
print(e.body)
|
155
|
+
```
|
156
|
+
|
157
|
+
## Streaming
|
158
|
+
|
159
|
+
The SDK supports streaming responses, as well, the response will be a generator that you can loop over.
|
160
|
+
|
161
|
+
```python
|
162
|
+
from cartesia import Cartesia
|
163
|
+
from cartesia.tts import Controls, OutputFormat_RawParams, TtsRequestIdSpecifierParams
|
164
|
+
|
165
|
+
client = Cartesia(
|
166
|
+
api_key="YOUR_API_KEY",
|
167
|
+
)
|
168
|
+
response = client.tts.sse(
|
169
|
+
model_id="string",
|
170
|
+
transcript="string",
|
171
|
+
voice={
|
172
|
+
"id": "string",
|
173
|
+
"experimental_controls": {
|
174
|
+
speed=1.1,
|
175
|
+
emotion="anger:lowest",
|
176
|
+
},
|
177
|
+
},
|
178
|
+
language="en",
|
179
|
+
output_format={},
|
180
|
+
duration=1.1,
|
181
|
+
)
|
182
|
+
for chunk in response:
|
183
|
+
yield chunk
|
184
|
+
```
|
185
|
+
|
186
|
+
## WebSocket
|
187
|
+
|
188
|
+
```python
|
189
|
+
from cartesia import Cartesia
|
190
|
+
from cartesia.tts import TtsRequestEmbeddingSpecifierParams, OutputFormat_RawParams
|
191
|
+
import pyaudio
|
192
|
+
|
193
|
+
client = Cartesia(
|
194
|
+
api_key="YOUR_API_KEY",
|
195
|
+
)
|
196
|
+
voice_id = "a0e99841-438c-4a64-b679-ae501e7d6091"
|
197
|
+
voice = client.voices.get(id=voice_id)
|
198
|
+
transcript = "Hello! Welcome to Cartesia"
|
199
|
+
|
200
|
+
# You can check out our models at https://docs.cartesia.ai/getting-started/available-models
|
201
|
+
model_id = "sonic-english"
|
202
|
+
|
203
|
+
p = pyaudio.PyAudio()
|
204
|
+
rate = 22050
|
205
|
+
|
206
|
+
stream = None
|
207
|
+
|
208
|
+
# Set up the websocket connection
|
209
|
+
ws = client.tts.websocket()
|
210
|
+
|
211
|
+
# Generate and stream audio using the websocket
|
212
|
+
for output in ws.send(
|
213
|
+
model_id=model_id,
|
214
|
+
transcript=transcript,
|
215
|
+
voice={"embedding": voice.embedding},
|
216
|
+
stream=True,
|
217
|
+
output_format={
|
218
|
+
"container": "raw",
|
219
|
+
"encoding": "pcm_f32le",
|
220
|
+
"sample_rate": 22050
|
221
|
+
},
|
222
|
+
):
|
223
|
+
buffer = output.audio
|
224
|
+
|
225
|
+
if not stream:
|
226
|
+
stream = p.open(format=pyaudio.paFloat32, channels=1, rate=rate, output=True)
|
227
|
+
|
228
|
+
# Write the audio data to the stream
|
229
|
+
stream.write(buffer)
|
230
|
+
|
231
|
+
stream.stop_stream()
|
232
|
+
stream.close()
|
233
|
+
p.terminate()
|
234
|
+
|
235
|
+
ws.close() # Close the websocket connection
|
236
|
+
```
|
237
|
+
|
238
|
+
## Advanced
|
239
|
+
|
240
|
+
### Retries
|
241
|
+
|
242
|
+
The SDK is instrumented with automatic retries with exponential backoff. A request will be retried as long
|
243
|
+
as the request is deemed retriable and the number of retry attempts has not grown larger than the configured
|
244
|
+
retry limit (default: 2).
|
245
|
+
|
246
|
+
A request is deemed retriable when any of the following HTTP status codes is returned:
|
247
|
+
|
248
|
+
- [408](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/408) (Timeout)
|
249
|
+
- [429](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/429) (Too Many Requests)
|
250
|
+
- [5XX](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/500) (Internal Server Errors)
|
251
|
+
|
252
|
+
Use the `max_retries` request option to configure this behavior.
|
253
|
+
|
254
|
+
```python
|
255
|
+
client.tts.bytes(..., request_options={
|
256
|
+
"max_retries": 1
|
257
|
+
})
|
258
|
+
```
|
259
|
+
|
260
|
+
### Timeouts
|
261
|
+
|
262
|
+
The SDK defaults to a 60 second timeout. You can configure this with a timeout option at the client or request level.
|
263
|
+
|
264
|
+
```python
|
265
|
+
|
266
|
+
from cartesia import Cartesia
|
267
|
+
|
268
|
+
client = Cartesia(
|
269
|
+
...,
|
270
|
+
timeout=20.0,
|
271
|
+
)
|
272
|
+
|
273
|
+
|
274
|
+
# Override timeout for a specific method
|
275
|
+
client.tts.bytes(..., request_options={
|
276
|
+
"timeout_in_seconds": 1
|
277
|
+
})
|
278
|
+
```
|
279
|
+
|
280
|
+
### Custom Client
|
281
|
+
|
282
|
+
You can override the `httpx` client to customize it for your use-case. Some common use-cases include support for proxies
|
283
|
+
and transports.
|
284
|
+
```python
|
285
|
+
import httpx
|
286
|
+
from cartesia import Cartesia
|
287
|
+
|
288
|
+
client = Cartesia(
|
289
|
+
...,
|
290
|
+
httpx_client=httpx.Client(
|
291
|
+
proxies="http://my.test.proxy.example.com",
|
292
|
+
transport=httpx.HTTPTransport(local_address="0.0.0.0"),
|
293
|
+
),
|
294
|
+
)
|
295
|
+
```
|
296
|
+
|
297
|
+
## Contributing
|
298
|
+
|
299
|
+
While we value open-source contributions to this SDK, this library is generated programmatically.
|
300
|
+
Additions made directly to this library would have to be moved over to our generation code,
|
301
|
+
otherwise they would be overwritten upon the next generated release. Feel free to open a PR as
|
302
|
+
a proof of concept, but know that we will not be able to merge it as-is. We suggest opening
|
303
|
+
an issue first to discuss with us!
|
304
|
+
|
305
|
+
On the other hand, contributions to the README are always very welcome!
|
306
|
+
|
@@ -0,0 +1,274 @@
|
|
1
|
+
# Cartesia Python Library
|
2
|
+
|
3
|
+
[](https://buildwithfern.com?utm_source=github&utm_medium=github&utm_campaign=readme&utm_source=https%3A%2F%2Fgithub.com%2Fcartesia-ai%2Fcartesia-python)
|
4
|
+
[](https://pypi.python.org/pypi/cartesia)
|
5
|
+
|
6
|
+
The Cartesia Python library provides convenient access to the Cartesia API from Python.
|
7
|
+
|
8
|
+
## Documentation
|
9
|
+
|
10
|
+
Our complete API documentation can be found [on docs.cartesia.ai](https://docs.cartesia.ai).
|
11
|
+
|
12
|
+
## Installation
|
13
|
+
|
14
|
+
```sh
|
15
|
+
pip install cartesia
|
16
|
+
```
|
17
|
+
|
18
|
+
## Reference
|
19
|
+
|
20
|
+
A full reference for this library is available [here](./reference.md).
|
21
|
+
|
22
|
+
## Voices
|
23
|
+
|
24
|
+
```python
|
25
|
+
from cartesia import Cartesia
|
26
|
+
import os
|
27
|
+
|
28
|
+
client = Cartesia(api_key=os.environ.get("CARTESIA_API_KEY"))
|
29
|
+
|
30
|
+
# Get all available voices
|
31
|
+
voices = client.voices.list()
|
32
|
+
print(voices)
|
33
|
+
|
34
|
+
# Get a specific voice
|
35
|
+
voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
|
36
|
+
print("The embedding for", voice["name"], "is", voice["embedding"])
|
37
|
+
|
38
|
+
# Clone a voice using filepath
|
39
|
+
cloned_voice_embedding = client.voices.clone(filepath="path/to/voice")
|
40
|
+
|
41
|
+
# Mix voices together
|
42
|
+
mixed_voice_embedding = client.voices.mix(
|
43
|
+
[{ "id": "voice_id_1", "weight": 0.5 }, { "id": "voice_id_2", "weight": 0.25 }, { "id": "voice_id_3", "weight": 0.25 }]
|
44
|
+
)
|
45
|
+
|
46
|
+
# Create a new voice
|
47
|
+
new_voice = client.voices.create(
|
48
|
+
name="New Voice",
|
49
|
+
description="A clone of my own voice",
|
50
|
+
embedding=cloned_voice_embedding,
|
51
|
+
)
|
52
|
+
```
|
53
|
+
|
54
|
+
## Usage
|
55
|
+
|
56
|
+
Instantiate and use the client with the following:
|
57
|
+
|
58
|
+
```python
|
59
|
+
from cartesia import Cartesia
|
60
|
+
from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
|
61
|
+
|
62
|
+
client = Cartesia(
|
63
|
+
api_key="YOUR_API_KEY",
|
64
|
+
)
|
65
|
+
client.tts.bytes(
|
66
|
+
model_id="sonic-english",
|
67
|
+
transcript="Hello, world!",
|
68
|
+
voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
|
69
|
+
),
|
70
|
+
language="en",
|
71
|
+
output_format={
|
72
|
+
"container": "raw",
|
73
|
+
"sample_rate": 44100,
|
74
|
+
"encoding": "pcm_f32le",
|
75
|
+
},
|
76
|
+
)
|
77
|
+
```
|
78
|
+
|
79
|
+
## Async Client
|
80
|
+
|
81
|
+
The SDK also exports an `async` client so that you can make non-blocking calls to our API.
|
82
|
+
|
83
|
+
```python
|
84
|
+
import asyncio
|
85
|
+
|
86
|
+
from cartesia import AsyncCartesia
|
87
|
+
from cartesia.tts import OutputFormat_Raw, TtsRequestIdSpecifier
|
88
|
+
|
89
|
+
client = AsyncCartesia(
|
90
|
+
api_key="YOUR_API_KEY",
|
91
|
+
)
|
92
|
+
|
93
|
+
|
94
|
+
async def main() -> None:
|
95
|
+
await client.tts.bytes(
|
96
|
+
model_id="sonic-english",
|
97
|
+
transcript="Hello, world!",
|
98
|
+
voice={"id": "694f9389-aac1-45b6-b726-9d9369183238"},
|
99
|
+
language="en",
|
100
|
+
output_format={
|
101
|
+
"container": "raw",
|
102
|
+
"sample_rate": 44100,
|
103
|
+
"encoding": "pcm_f32le",
|
104
|
+
},
|
105
|
+
)
|
106
|
+
|
107
|
+
|
108
|
+
asyncio.run(main())
|
109
|
+
```
|
110
|
+
|
111
|
+
## Exception Handling
|
112
|
+
|
113
|
+
When the API returns a non-success status code (4xx or 5xx response), a subclass of the following error
|
114
|
+
will be thrown.
|
115
|
+
|
116
|
+
```python
|
117
|
+
from cartesia.core.api_error import ApiError
|
118
|
+
|
119
|
+
try:
|
120
|
+
client.tts.bytes(...)
|
121
|
+
except ApiError as e:
|
122
|
+
print(e.status_code)
|
123
|
+
print(e.body)
|
124
|
+
```
|
125
|
+
|
126
|
+
## Streaming
|
127
|
+
|
128
|
+
The SDK supports streaming responses, as well, the response will be a generator that you can loop over.
|
129
|
+
|
130
|
+
```python
|
131
|
+
from cartesia import Cartesia
|
132
|
+
from cartesia.tts import Controls, OutputFormat_RawParams, TtsRequestIdSpecifierParams
|
133
|
+
|
134
|
+
client = Cartesia(
|
135
|
+
api_key="YOUR_API_KEY",
|
136
|
+
)
|
137
|
+
response = client.tts.sse(
|
138
|
+
model_id="string",
|
139
|
+
transcript="string",
|
140
|
+
voice={
|
141
|
+
"id": "string",
|
142
|
+
"experimental_controls": {
|
143
|
+
speed=1.1,
|
144
|
+
emotion="anger:lowest",
|
145
|
+
},
|
146
|
+
},
|
147
|
+
language="en",
|
148
|
+
output_format={},
|
149
|
+
duration=1.1,
|
150
|
+
)
|
151
|
+
for chunk in response:
|
152
|
+
yield chunk
|
153
|
+
```
|
154
|
+
|
155
|
+
## WebSocket
|
156
|
+
|
157
|
+
```python
|
158
|
+
from cartesia import Cartesia
|
159
|
+
from cartesia.tts import TtsRequestEmbeddingSpecifierParams, OutputFormat_RawParams
|
160
|
+
import pyaudio
|
161
|
+
|
162
|
+
client = Cartesia(
|
163
|
+
api_key="YOUR_API_KEY",
|
164
|
+
)
|
165
|
+
voice_id = "a0e99841-438c-4a64-b679-ae501e7d6091"
|
166
|
+
voice = client.voices.get(id=voice_id)
|
167
|
+
transcript = "Hello! Welcome to Cartesia"
|
168
|
+
|
169
|
+
# You can check out our models at https://docs.cartesia.ai/getting-started/available-models
|
170
|
+
model_id = "sonic-english"
|
171
|
+
|
172
|
+
p = pyaudio.PyAudio()
|
173
|
+
rate = 22050
|
174
|
+
|
175
|
+
stream = None
|
176
|
+
|
177
|
+
# Set up the websocket connection
|
178
|
+
ws = client.tts.websocket()
|
179
|
+
|
180
|
+
# Generate and stream audio using the websocket
|
181
|
+
for output in ws.send(
|
182
|
+
model_id=model_id,
|
183
|
+
transcript=transcript,
|
184
|
+
voice={"embedding": voice.embedding},
|
185
|
+
stream=True,
|
186
|
+
output_format={
|
187
|
+
"container": "raw",
|
188
|
+
"encoding": "pcm_f32le",
|
189
|
+
"sample_rate": 22050
|
190
|
+
},
|
191
|
+
):
|
192
|
+
buffer = output.audio
|
193
|
+
|
194
|
+
if not stream:
|
195
|
+
stream = p.open(format=pyaudio.paFloat32, channels=1, rate=rate, output=True)
|
196
|
+
|
197
|
+
# Write the audio data to the stream
|
198
|
+
stream.write(buffer)
|
199
|
+
|
200
|
+
stream.stop_stream()
|
201
|
+
stream.close()
|
202
|
+
p.terminate()
|
203
|
+
|
204
|
+
ws.close() # Close the websocket connection
|
205
|
+
```
|
206
|
+
|
207
|
+
## Advanced
|
208
|
+
|
209
|
+
### Retries
|
210
|
+
|
211
|
+
The SDK is instrumented with automatic retries with exponential backoff. A request will be retried as long
|
212
|
+
as the request is deemed retriable and the number of retry attempts has not grown larger than the configured
|
213
|
+
retry limit (default: 2).
|
214
|
+
|
215
|
+
A request is deemed retriable when any of the following HTTP status codes is returned:
|
216
|
+
|
217
|
+
- [408](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/408) (Timeout)
|
218
|
+
- [429](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/429) (Too Many Requests)
|
219
|
+
- [5XX](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/500) (Internal Server Errors)
|
220
|
+
|
221
|
+
Use the `max_retries` request option to configure this behavior.
|
222
|
+
|
223
|
+
```python
|
224
|
+
client.tts.bytes(..., request_options={
|
225
|
+
"max_retries": 1
|
226
|
+
})
|
227
|
+
```
|
228
|
+
|
229
|
+
### Timeouts
|
230
|
+
|
231
|
+
The SDK defaults to a 60 second timeout. You can configure this with a timeout option at the client or request level.
|
232
|
+
|
233
|
+
```python
|
234
|
+
|
235
|
+
from cartesia import Cartesia
|
236
|
+
|
237
|
+
client = Cartesia(
|
238
|
+
...,
|
239
|
+
timeout=20.0,
|
240
|
+
)
|
241
|
+
|
242
|
+
|
243
|
+
# Override timeout for a specific method
|
244
|
+
client.tts.bytes(..., request_options={
|
245
|
+
"timeout_in_seconds": 1
|
246
|
+
})
|
247
|
+
```
|
248
|
+
|
249
|
+
### Custom Client
|
250
|
+
|
251
|
+
You can override the `httpx` client to customize it for your use-case. Some common use-cases include support for proxies
|
252
|
+
and transports.
|
253
|
+
```python
|
254
|
+
import httpx
|
255
|
+
from cartesia import Cartesia
|
256
|
+
|
257
|
+
client = Cartesia(
|
258
|
+
...,
|
259
|
+
httpx_client=httpx.Client(
|
260
|
+
proxies="http://my.test.proxy.example.com",
|
261
|
+
transport=httpx.HTTPTransport(local_address="0.0.0.0"),
|
262
|
+
),
|
263
|
+
)
|
264
|
+
```
|
265
|
+
|
266
|
+
## Contributing
|
267
|
+
|
268
|
+
While we value open-source contributions to this SDK, this library is generated programmatically.
|
269
|
+
Additions made directly to this library would have to be moved over to our generation code,
|
270
|
+
otherwise they would be overwritten upon the next generated release. Feel free to open a PR as
|
271
|
+
a proof of concept, but know that we will not be able to merge it as-is. We suggest opening
|
272
|
+
an issue first to discuss with us!
|
273
|
+
|
274
|
+
On the other hand, contributions to the README are always very welcome!
|
@@ -0,0 +1,70 @@
|
|
1
|
+
[project]
|
2
|
+
name = "cartesia"
|
3
|
+
|
4
|
+
[tool.poetry]
|
5
|
+
name = "cartesia"
|
6
|
+
version = "2.0.0a0"
|
7
|
+
description = ""
|
8
|
+
readme = "README.md"
|
9
|
+
authors = []
|
10
|
+
keywords = []
|
11
|
+
|
12
|
+
classifiers = [
|
13
|
+
"Intended Audience :: Developers",
|
14
|
+
"Programming Language :: Python",
|
15
|
+
"Programming Language :: Python :: 3",
|
16
|
+
"Programming Language :: Python :: 3.8",
|
17
|
+
"Programming Language :: Python :: 3.9",
|
18
|
+
"Programming Language :: Python :: 3.10",
|
19
|
+
"Programming Language :: Python :: 3.11",
|
20
|
+
"Programming Language :: Python :: 3.12",
|
21
|
+
"Operating System :: OS Independent",
|
22
|
+
"Operating System :: POSIX",
|
23
|
+
"Operating System :: MacOS",
|
24
|
+
"Operating System :: POSIX :: Linux",
|
25
|
+
"Operating System :: Microsoft :: Windows",
|
26
|
+
"Topic :: Software Development :: Libraries :: Python Modules",
|
27
|
+
"Typing :: Typed"
|
28
|
+
]
|
29
|
+
packages = [
|
30
|
+
{ include = "cartesia", from = "src"}
|
31
|
+
]
|
32
|
+
|
33
|
+
[project.urls]
|
34
|
+
Repository = 'https://github.com/cartesia-ai/cartesia-python'
|
35
|
+
|
36
|
+
[tool.poetry.dependencies]
|
37
|
+
python = "^3.8"
|
38
|
+
aiohttp = ">=3.10.10"
|
39
|
+
httpx = ">=0.21.2"
|
40
|
+
httpx-sse = "0.4.0"
|
41
|
+
iterators = ">=0.2.0"
|
42
|
+
pydantic = ">= 1.9.2"
|
43
|
+
pydantic-core = "^2.18.2"
|
44
|
+
pydub = ">=0.25.1"
|
45
|
+
typing_extensions = ">= 4.0.0"
|
46
|
+
websockets = ">=10.4"
|
47
|
+
|
48
|
+
[tool.poetry.dev-dependencies]
|
49
|
+
mypy = "1.0.1"
|
50
|
+
pytest = "^7.4.0"
|
51
|
+
pytest-asyncio = "^0.23.5"
|
52
|
+
python-dateutil = "^2.9.0"
|
53
|
+
types-python-dateutil = "^2.9.0.20240316"
|
54
|
+
numpy = ">=1.2.1"
|
55
|
+
ruff = "^0.5.6"
|
56
|
+
|
57
|
+
[tool.pytest.ini_options]
|
58
|
+
testpaths = [ "tests" ]
|
59
|
+
asyncio_mode = "auto"
|
60
|
+
|
61
|
+
[tool.mypy]
|
62
|
+
plugins = ["pydantic.mypy"]
|
63
|
+
|
64
|
+
[tool.ruff]
|
65
|
+
line-length = 120
|
66
|
+
|
67
|
+
|
68
|
+
[build-system]
|
69
|
+
requires = ["poetry-core"]
|
70
|
+
build-backend = "poetry.core.masonry.api"
|