cartesia 1.3.1__py3-none-any.whl → 2.0.0__py3-none-any.whl
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/__init__.py +302 -3
- cartesia/api_status/__init__.py +6 -0
- cartesia/api_status/client.py +104 -0
- cartesia/api_status/requests/__init__.py +5 -0
- cartesia/api_status/requests/api_info.py +8 -0
- cartesia/api_status/types/__init__.py +5 -0
- cartesia/api_status/types/api_info.py +20 -0
- cartesia/base_client.py +156 -0
- cartesia/client.py +163 -40
- cartesia/core/__init__.py +50 -0
- cartesia/core/api_error.py +15 -0
- cartesia/core/client_wrapper.py +55 -0
- cartesia/core/datetime_utils.py +28 -0
- cartesia/core/file.py +67 -0
- cartesia/core/http_client.py +499 -0
- cartesia/core/jsonable_encoder.py +101 -0
- cartesia/core/pagination.py +88 -0
- cartesia/core/pydantic_utilities.py +296 -0
- cartesia/core/query_encoder.py +58 -0
- cartesia/core/remove_none_from_dict.py +11 -0
- cartesia/core/request_options.py +35 -0
- cartesia/core/serialization.py +272 -0
- cartesia/datasets/__init__.py +24 -0
- cartesia/datasets/requests/__init__.py +15 -0
- cartesia/datasets/requests/create_dataset_request.py +7 -0
- cartesia/datasets/requests/dataset.py +9 -0
- cartesia/datasets/requests/dataset_file.py +9 -0
- cartesia/datasets/requests/paginated_dataset_files.py +10 -0
- cartesia/datasets/requests/paginated_datasets.py +10 -0
- cartesia/datasets/types/__init__.py +17 -0
- cartesia/datasets/types/create_dataset_request.py +19 -0
- cartesia/datasets/types/dataset.py +21 -0
- cartesia/datasets/types/dataset_file.py +21 -0
- cartesia/datasets/types/file_purpose.py +5 -0
- cartesia/datasets/types/paginated_dataset_files.py +21 -0
- cartesia/datasets/types/paginated_datasets.py +21 -0
- cartesia/embedding/__init__.py +5 -0
- cartesia/embedding/types/__init__.py +5 -0
- cartesia/embedding/types/embedding.py +201 -0
- cartesia/environment.py +7 -0
- cartesia/infill/__init__.py +2 -0
- cartesia/infill/client.py +318 -0
- cartesia/tts/__init__.py +167 -0
- cartesia/{_async_websocket.py → tts/_async_websocket.py} +212 -85
- cartesia/tts/_websocket.py +479 -0
- cartesia/tts/client.py +407 -0
- cartesia/tts/requests/__init__.py +76 -0
- cartesia/tts/requests/cancel_context_request.py +17 -0
- cartesia/tts/requests/controls.py +11 -0
- cartesia/tts/requests/generation_request.py +58 -0
- cartesia/tts/requests/mp_3_output_format.py +11 -0
- cartesia/tts/requests/output_format.py +30 -0
- cartesia/tts/requests/phoneme_timestamps.py +10 -0
- cartesia/tts/requests/raw_output_format.py +11 -0
- cartesia/tts/requests/speed.py +7 -0
- cartesia/tts/requests/tts_request.py +24 -0
- cartesia/tts/requests/tts_request_embedding_specifier.py +16 -0
- cartesia/tts/requests/tts_request_id_specifier.py +16 -0
- cartesia/tts/requests/tts_request_voice_specifier.py +7 -0
- cartesia/tts/requests/wav_output_format.py +7 -0
- cartesia/tts/requests/web_socket_base_response.py +11 -0
- cartesia/tts/requests/web_socket_chunk_response.py +11 -0
- cartesia/tts/requests/web_socket_done_response.py +7 -0
- cartesia/tts/requests/web_socket_error_response.py +7 -0
- cartesia/tts/requests/web_socket_flush_done_response.py +9 -0
- cartesia/tts/requests/web_socket_phoneme_timestamps_response.py +9 -0
- cartesia/tts/requests/web_socket_raw_output_format.py +11 -0
- cartesia/tts/requests/web_socket_request.py +7 -0
- cartesia/tts/requests/web_socket_response.py +70 -0
- cartesia/tts/requests/web_socket_stream_options.py +8 -0
- cartesia/tts/requests/web_socket_timestamps_response.py +9 -0
- cartesia/tts/requests/web_socket_tts_output.py +18 -0
- cartesia/tts/requests/web_socket_tts_request.py +25 -0
- cartesia/tts/requests/word_timestamps.py +10 -0
- cartesia/tts/socket_client.py +302 -0
- cartesia/tts/types/__init__.py +90 -0
- cartesia/tts/types/cancel_context_request.py +28 -0
- cartesia/tts/types/context_id.py +3 -0
- cartesia/tts/types/controls.py +22 -0
- cartesia/tts/types/emotion.py +34 -0
- cartesia/tts/types/flush_id.py +3 -0
- cartesia/tts/types/generation_request.py +71 -0
- cartesia/tts/types/mp_3_output_format.py +23 -0
- cartesia/tts/types/natural_specifier.py +5 -0
- cartesia/tts/types/numerical_specifier.py +3 -0
- cartesia/tts/types/output_format.py +58 -0
- cartesia/tts/types/phoneme_timestamps.py +21 -0
- cartesia/tts/types/raw_encoding.py +5 -0
- cartesia/tts/types/raw_output_format.py +22 -0
- cartesia/tts/types/speed.py +7 -0
- cartesia/tts/types/supported_language.py +7 -0
- cartesia/tts/types/tts_request.py +35 -0
- cartesia/tts/types/tts_request_embedding_specifier.py +27 -0
- cartesia/tts/types/tts_request_id_specifier.py +27 -0
- cartesia/tts/types/tts_request_voice_specifier.py +7 -0
- cartesia/tts/types/wav_output_format.py +17 -0
- cartesia/tts/types/web_socket_base_response.py +22 -0
- cartesia/tts/types/web_socket_chunk_response.py +22 -0
- cartesia/tts/types/web_socket_done_response.py +17 -0
- cartesia/tts/types/web_socket_error_response.py +19 -0
- cartesia/tts/types/web_socket_flush_done_response.py +21 -0
- cartesia/tts/types/web_socket_phoneme_timestamps_response.py +20 -0
- cartesia/tts/types/web_socket_raw_output_format.py +22 -0
- cartesia/tts/types/web_socket_request.py +7 -0
- cartesia/tts/types/web_socket_response.py +125 -0
- cartesia/tts/types/web_socket_stream_options.py +19 -0
- cartesia/tts/types/web_socket_timestamps_response.py +20 -0
- cartesia/tts/types/web_socket_tts_output.py +29 -0
- cartesia/tts/types/web_socket_tts_request.py +37 -0
- cartesia/tts/types/word_timestamps.py +21 -0
- cartesia/{_constants.py → tts/utils/constants.py} +2 -2
- cartesia/tts/utils/tts.py +64 -0
- cartesia/tts/utils/types.py +70 -0
- cartesia/version.py +3 -1
- cartesia/voice_changer/__init__.py +27 -0
- cartesia/voice_changer/client.py +395 -0
- cartesia/voice_changer/requests/__init__.py +15 -0
- cartesia/voice_changer/requests/streaming_response.py +38 -0
- cartesia/voice_changer/types/__init__.py +17 -0
- cartesia/voice_changer/types/output_format_container.py +5 -0
- cartesia/voice_changer/types/streaming_response.py +64 -0
- cartesia/voices/__init__.py +81 -0
- cartesia/voices/client.py +1218 -0
- cartesia/voices/requests/__init__.py +29 -0
- cartesia/voices/requests/create_voice_request.py +23 -0
- cartesia/voices/requests/embedding_response.py +8 -0
- cartesia/voices/requests/embedding_specifier.py +10 -0
- cartesia/voices/requests/get_voices_response.py +24 -0
- cartesia/voices/requests/id_specifier.py +10 -0
- cartesia/voices/requests/localize_dialect.py +11 -0
- cartesia/voices/requests/localize_voice_request.py +28 -0
- cartesia/voices/requests/mix_voice_specifier.py +7 -0
- cartesia/voices/requests/mix_voices_request.py +9 -0
- cartesia/voices/requests/update_voice_request.py +15 -0
- cartesia/voices/requests/voice.py +43 -0
- cartesia/voices/requests/voice_metadata.py +36 -0
- cartesia/voices/types/__init__.py +53 -0
- cartesia/voices/types/base_voice_id.py +5 -0
- cartesia/voices/types/clone_mode.py +5 -0
- cartesia/voices/types/create_voice_request.py +34 -0
- cartesia/voices/types/embedding_response.py +20 -0
- cartesia/voices/types/embedding_specifier.py +22 -0
- cartesia/voices/types/gender.py +5 -0
- cartesia/voices/types/gender_presentation.py +5 -0
- cartesia/voices/types/get_voices_response.py +34 -0
- cartesia/voices/types/id_specifier.py +22 -0
- cartesia/voices/types/localize_dialect.py +11 -0
- cartesia/voices/types/localize_english_dialect.py +5 -0
- cartesia/voices/types/localize_french_dialect.py +5 -0
- cartesia/voices/types/localize_portuguese_dialect.py +5 -0
- cartesia/voices/types/localize_spanish_dialect.py +5 -0
- cartesia/voices/types/localize_target_language.py +7 -0
- cartesia/voices/types/localize_voice_request.py +39 -0
- cartesia/voices/types/mix_voice_specifier.py +7 -0
- cartesia/voices/types/mix_voices_request.py +20 -0
- cartesia/voices/types/update_voice_request.py +27 -0
- cartesia/voices/types/voice.py +54 -0
- cartesia/voices/types/voice_expand_options.py +5 -0
- cartesia/voices/types/voice_id.py +3 -0
- cartesia/voices/types/voice_metadata.py +48 -0
- cartesia/voices/types/weight.py +3 -0
- cartesia-2.0.0.dist-info/METADATA +414 -0
- cartesia-2.0.0.dist-info/RECORD +165 -0
- {cartesia-1.3.1.dist-info → cartesia-2.0.0.dist-info}/WHEEL +1 -1
- cartesia/_async_sse.py +0 -95
- cartesia/_logger.py +0 -3
- cartesia/_sse.py +0 -143
- cartesia/_types.py +0 -70
- cartesia/_websocket.py +0 -358
- cartesia/async_client.py +0 -82
- cartesia/async_tts.py +0 -63
- cartesia/resource.py +0 -44
- cartesia/tts.py +0 -137
- cartesia/utils/deprecated.py +0 -55
- cartesia/utils/retry.py +0 -87
- cartesia/utils/tts.py +0 -78
- cartesia/voices.py +0 -208
- cartesia-1.3.1.dist-info/METADATA +0 -661
- cartesia-1.3.1.dist-info/RECORD +0 -23
- cartesia-1.3.1.dist-info/licenses/LICENSE.md +0 -21
- /cartesia/{utils/__init__.py → py.typed} +0 -0
cartesia/utils/deprecated.py
DELETED
@@ -1,55 +0,0 @@
|
|
1
|
-
import os
|
2
|
-
import warnings
|
3
|
-
from typing import Any, Callable, TypeVar
|
4
|
-
|
5
|
-
TCallable = TypeVar("TCallable", bound=Callable[..., Any])
|
6
|
-
|
7
|
-
# List of statistics of deprecated functions.
|
8
|
-
# This should only be used by the test suite to find any deprecated functions
|
9
|
-
# that should be removed for this version.
|
10
|
-
_TRACK_DEPRECATED_FUNCTION_STATS = os.environ.get("CARTESIA_TEST_DEPRECATED", "").lower() == "true"
|
11
|
-
_DEPRECATED_FUNCTION_STATS = []
|
12
|
-
|
13
|
-
|
14
|
-
def deprecated(
|
15
|
-
reason=None, vdeprecated=None, vremove=None, replacement=None
|
16
|
-
) -> Callable[[TCallable], TCallable]:
|
17
|
-
local_vars = locals()
|
18
|
-
|
19
|
-
def fn(func: TCallable) -> TCallable:
|
20
|
-
if isinstance(func, classmethod):
|
21
|
-
func = func.__func__
|
22
|
-
msg = _get_deprecated_msg(func, reason, vdeprecated, vremove, replacement)
|
23
|
-
warnings.warn(msg, DeprecationWarning)
|
24
|
-
return func
|
25
|
-
|
26
|
-
if _TRACK_DEPRECATED_FUNCTION_STATS: # pragma: no cover
|
27
|
-
_DEPRECATED_FUNCTION_STATS.append(local_vars)
|
28
|
-
|
29
|
-
return fn
|
30
|
-
|
31
|
-
|
32
|
-
def _get_deprecated_msg(wrapped, reason, vdeprecated, vremoved, replacement=None):
|
33
|
-
fmt = "{name} is deprecated"
|
34
|
-
if vdeprecated:
|
35
|
-
fmt += " since v{vdeprecated}"
|
36
|
-
if vremoved:
|
37
|
-
fmt += " and will be removed in v{vremoved}"
|
38
|
-
fmt += "."
|
39
|
-
|
40
|
-
if reason:
|
41
|
-
fmt += " ({reason})"
|
42
|
-
if replacement:
|
43
|
-
fmt += " -- Use {replacement} instead."
|
44
|
-
|
45
|
-
return fmt.format(
|
46
|
-
name=wrapped.__name__,
|
47
|
-
reason=reason or "",
|
48
|
-
vdeprecated=vdeprecated or "",
|
49
|
-
vremoved=vremoved or "",
|
50
|
-
replacement=replacement or "",
|
51
|
-
)
|
52
|
-
|
53
|
-
|
54
|
-
# This method is taken from the following source:
|
55
|
-
# https://github.com/ad12/meddlr/blob/main/meddlr/utils/deprecated.py
|
cartesia/utils/retry.py
DELETED
@@ -1,87 +0,0 @@
|
|
1
|
-
import asyncio
|
2
|
-
import time
|
3
|
-
from functools import wraps
|
4
|
-
from http.client import RemoteDisconnected
|
5
|
-
|
6
|
-
from aiohttp.client_exceptions import ServerDisconnectedError
|
7
|
-
from httpx import TimeoutException
|
8
|
-
from requests.exceptions import ConnectionError
|
9
|
-
|
10
|
-
|
11
|
-
def retry_on_connection_error(max_retries=3, backoff_factor=1, logger=None):
|
12
|
-
"""Retry a function if a ConnectionError, RemoteDisconnected, ServerDisconnectedError, or TimeoutException occurs.
|
13
|
-
|
14
|
-
Args:
|
15
|
-
max_retries (int): The maximum number of retries.
|
16
|
-
backoff_factor (int): The factor to increase the delay between retries.
|
17
|
-
logger (logging.Logger): The logger to use for logging.
|
18
|
-
"""
|
19
|
-
|
20
|
-
def decorator(func):
|
21
|
-
@wraps(func)
|
22
|
-
def wrapper(*args, **kwargs):
|
23
|
-
retry_count = 0
|
24
|
-
while retry_count < max_retries:
|
25
|
-
try:
|
26
|
-
return func(*args, **kwargs)
|
27
|
-
except (
|
28
|
-
ConnectionError,
|
29
|
-
RemoteDisconnected,
|
30
|
-
ServerDisconnectedError,
|
31
|
-
TimeoutException,
|
32
|
-
) as e:
|
33
|
-
logger.info(f"Retrying after exception: {e}")
|
34
|
-
retry_count += 1
|
35
|
-
if retry_count < max_retries:
|
36
|
-
delay = backoff_factor * (2 ** (retry_count - 1))
|
37
|
-
logger.warn(
|
38
|
-
f"Attempt {retry_count + 1}/{max_retries} in {delay} seconds..."
|
39
|
-
)
|
40
|
-
time.sleep(delay)
|
41
|
-
else:
|
42
|
-
raise Exception(f"Exception occurred after {max_retries} tries.") from e
|
43
|
-
|
44
|
-
return wrapper
|
45
|
-
|
46
|
-
return decorator
|
47
|
-
|
48
|
-
|
49
|
-
def retry_on_connection_error_async(max_retries=3, backoff_factor=1, logger=None):
|
50
|
-
"""Retry an asynchronous function if a ConnectionError, RemoteDisconnected, ServerDisconnectedError, or TimeoutException occurs.
|
51
|
-
|
52
|
-
Args:
|
53
|
-
max_retries (int): The maximum number of retries.
|
54
|
-
backoff_factor (int): The factor to increase the delay between retries.
|
55
|
-
logger (logging.Logger): The logger to use for logging.
|
56
|
-
"""
|
57
|
-
|
58
|
-
def decorator(func):
|
59
|
-
@wraps(func)
|
60
|
-
async def wrapper(*args, **kwargs):
|
61
|
-
retry_count = 0
|
62
|
-
while retry_count < max_retries:
|
63
|
-
try:
|
64
|
-
async for chunk in func(*args, **kwargs):
|
65
|
-
yield chunk
|
66
|
-
# If the function completes without raising an exception return
|
67
|
-
return
|
68
|
-
except (
|
69
|
-
ConnectionError,
|
70
|
-
RemoteDisconnected,
|
71
|
-
ServerDisconnectedError,
|
72
|
-
TimeoutException,
|
73
|
-
) as e:
|
74
|
-
logger.info(f"Retrying after exception: {e}")
|
75
|
-
retry_count += 1
|
76
|
-
if retry_count < max_retries:
|
77
|
-
delay = backoff_factor * (2 ** (retry_count - 1))
|
78
|
-
logger.warn(
|
79
|
-
f"Attempt {retry_count + 1}/{max_retries} in {delay} seconds..."
|
80
|
-
)
|
81
|
-
await asyncio.sleep(delay)
|
82
|
-
else:
|
83
|
-
raise Exception(f"Exception occurred after {max_retries} tries.") from e
|
84
|
-
|
85
|
-
return wrapper
|
86
|
-
|
87
|
-
return decorator
|
cartesia/utils/tts.py
DELETED
@@ -1,78 +0,0 @@
|
|
1
|
-
from typing import List, Optional
|
2
|
-
|
3
|
-
from cartesia._types import OutputFormat, VoiceControls
|
4
|
-
|
5
|
-
|
6
|
-
def _validate_and_construct_voice(
|
7
|
-
voice_id: Optional[str] = None,
|
8
|
-
voice_embedding: Optional[List[float]] = None,
|
9
|
-
experimental_voice_controls: Optional[VoiceControls] = None,
|
10
|
-
) -> dict:
|
11
|
-
if voice_id is None and voice_embedding is None:
|
12
|
-
raise ValueError("Either voice_id or voice_embedding must be specified.")
|
13
|
-
|
14
|
-
voice = {}
|
15
|
-
|
16
|
-
if voice_id is not None:
|
17
|
-
voice["id"] = voice_id
|
18
|
-
|
19
|
-
if voice_embedding is not None:
|
20
|
-
voice["embedding"] = voice_embedding
|
21
|
-
|
22
|
-
if experimental_voice_controls is not None:
|
23
|
-
voice["__experimental_controls"] = experimental_voice_controls
|
24
|
-
|
25
|
-
return voice
|
26
|
-
|
27
|
-
|
28
|
-
def _construct_tts_request(
|
29
|
-
*,
|
30
|
-
model_id: str,
|
31
|
-
output_format: OutputFormat,
|
32
|
-
transcript: Optional[str] = None,
|
33
|
-
voice_id: Optional[str] = None,
|
34
|
-
voice_embedding: Optional[List[float]] = None,
|
35
|
-
duration: Optional[int] = None,
|
36
|
-
language: Optional[str] = None,
|
37
|
-
add_timestamps: bool = False,
|
38
|
-
context_id: Optional[str] = None,
|
39
|
-
continue_: bool = False,
|
40
|
-
flush: bool = False,
|
41
|
-
_experimental_voice_controls: Optional[VoiceControls] = None,
|
42
|
-
):
|
43
|
-
tts_request = {
|
44
|
-
"model_id": model_id,
|
45
|
-
"voice": _validate_and_construct_voice(
|
46
|
-
voice_id,
|
47
|
-
voice_embedding=voice_embedding,
|
48
|
-
experimental_voice_controls=_experimental_voice_controls,
|
49
|
-
),
|
50
|
-
"output_format": {
|
51
|
-
"container": output_format["container"],
|
52
|
-
"encoding": output_format["encoding"],
|
53
|
-
"sample_rate": output_format["sample_rate"],
|
54
|
-
},
|
55
|
-
}
|
56
|
-
|
57
|
-
if language is not None:
|
58
|
-
tts_request["language"] = language
|
59
|
-
|
60
|
-
if transcript is not None:
|
61
|
-
tts_request["transcript"] = transcript
|
62
|
-
|
63
|
-
if duration is not None:
|
64
|
-
tts_request["duration"] = duration
|
65
|
-
|
66
|
-
if add_timestamps:
|
67
|
-
tts_request["add_timestamps"] = add_timestamps
|
68
|
-
|
69
|
-
if context_id is not None:
|
70
|
-
tts_request["context_id"] = context_id
|
71
|
-
|
72
|
-
if continue_:
|
73
|
-
tts_request["continue"] = continue_
|
74
|
-
|
75
|
-
if flush:
|
76
|
-
tts_request["flush"] = flush
|
77
|
-
|
78
|
-
return tts_request
|
cartesia/voices.py
DELETED
@@ -1,208 +0,0 @@
|
|
1
|
-
from typing import Dict, List, Optional, Union
|
2
|
-
|
3
|
-
import httpx
|
4
|
-
|
5
|
-
from cartesia._types import VoiceMetadata
|
6
|
-
from cartesia.resource import Resource
|
7
|
-
|
8
|
-
|
9
|
-
class Voices(Resource):
|
10
|
-
"""This resource contains methods to list, get, clone, and create voices in your Cartesia voice library.
|
11
|
-
|
12
|
-
Usage:
|
13
|
-
>>> client = Cartesia(api_key="your_api_key")
|
14
|
-
>>> voices = client.voices.list()
|
15
|
-
>>> voice = client.voices.get(id="a0e99841-438c-4a64-b679-ae501e7d6091")
|
16
|
-
>>> print("Voice Name:", voice["name"], "Voice Description:", voice["description"])
|
17
|
-
>>> embedding = client.voices.clone(filepath="path/to/clip.wav")
|
18
|
-
>>> new_voice = client.voices.create(
|
19
|
-
... name="My Voice", description="A new voice", embedding=embedding
|
20
|
-
... )
|
21
|
-
"""
|
22
|
-
|
23
|
-
def list(self) -> List[VoiceMetadata]:
|
24
|
-
"""List all voices in your voice library.
|
25
|
-
|
26
|
-
Returns:
|
27
|
-
This method returns a list of VoiceMetadata objects.
|
28
|
-
"""
|
29
|
-
response = httpx.get(
|
30
|
-
f"{self._http_url()}/voices",
|
31
|
-
headers=self.headers,
|
32
|
-
timeout=self.timeout,
|
33
|
-
)
|
34
|
-
|
35
|
-
if not response.is_success:
|
36
|
-
raise ValueError(f"Failed to get voices. Error: {response.text}")
|
37
|
-
|
38
|
-
voices = response.json()
|
39
|
-
return voices
|
40
|
-
|
41
|
-
def get(self, id: str) -> VoiceMetadata:
|
42
|
-
"""Get a voice by its ID.
|
43
|
-
|
44
|
-
Args:
|
45
|
-
id: The ID of the voice.
|
46
|
-
|
47
|
-
Returns:
|
48
|
-
A VoiceMetadata object containing the voice metadata.
|
49
|
-
"""
|
50
|
-
url = f"{self._http_url()}/voices/{id}"
|
51
|
-
response = httpx.get(url, headers=self.headers, timeout=self.timeout)
|
52
|
-
|
53
|
-
if not response.is_success:
|
54
|
-
raise ValueError(
|
55
|
-
f"Failed to get voice. Status Code: {response.status_code}\n"
|
56
|
-
f"Error: {response.text}"
|
57
|
-
)
|
58
|
-
|
59
|
-
return response.json()
|
60
|
-
|
61
|
-
def clone(
|
62
|
-
self,
|
63
|
-
filepath: Optional[str] = None,
|
64
|
-
enhance: str = True,
|
65
|
-
mode: str = "clip",
|
66
|
-
language: str = "en",
|
67
|
-
name: Optional[str] = None,
|
68
|
-
description: Optional[str] = None,
|
69
|
-
transcript: Optional[str] = None,
|
70
|
-
) -> Union[List[float], VoiceMetadata]:
|
71
|
-
"""Clone a voice from a clip.
|
72
|
-
|
73
|
-
Args:
|
74
|
-
filepath: The path to the clip file.
|
75
|
-
enhance: Whether to enhance the clip before cloning the voice (highly recommended). Defaults to True.
|
76
|
-
mode: The mode to use for cloning. Either "similarity" or "stability".
|
77
|
-
language: The language code of the language spoken in the clip. Defaults to "en".
|
78
|
-
name: The name of the cloned voice.
|
79
|
-
description: The description of the cloned voice.
|
80
|
-
transcript: The transcript of the clip. Only used if mode is "similarity".
|
81
|
-
|
82
|
-
Returns:
|
83
|
-
The embedding of the cloned voice as a list of floats.
|
84
|
-
"""
|
85
|
-
if not filepath:
|
86
|
-
raise ValueError("Filepath must be specified.")
|
87
|
-
headers = self.headers.copy()
|
88
|
-
headers.pop("Content-Type", None)
|
89
|
-
|
90
|
-
with open(filepath, "rb") as file:
|
91
|
-
files = {"clip": file}
|
92
|
-
data = {
|
93
|
-
"enhance": str(enhance).lower(),
|
94
|
-
"mode": mode,
|
95
|
-
}
|
96
|
-
if mode == "clip":
|
97
|
-
url = f"{self._http_url()}/voices/clone/clip"
|
98
|
-
response = httpx.post(
|
99
|
-
url, headers=headers, files=files, data=data, timeout=self.timeout
|
100
|
-
)
|
101
|
-
if not response.is_success:
|
102
|
-
raise ValueError(f"Failed to clone voice from clip. Error: {response.text}")
|
103
|
-
return response.json()["embedding"]
|
104
|
-
else:
|
105
|
-
data["name"] = name
|
106
|
-
data["description"] = description
|
107
|
-
data["language"] = language
|
108
|
-
if mode == "similarity" and transcript:
|
109
|
-
data["transcript"] = transcript
|
110
|
-
url = f"{self._http_url()}/voices/clone"
|
111
|
-
response = httpx.post(
|
112
|
-
url, headers=headers, files=files, data=data, timeout=self.timeout
|
113
|
-
)
|
114
|
-
if not response.is_success:
|
115
|
-
raise ValueError(
|
116
|
-
f"Failed to clone voice. Status Code: {response.status_code}\n"
|
117
|
-
f"Error: {response.text}"
|
118
|
-
)
|
119
|
-
return response.json()
|
120
|
-
|
121
|
-
def create(
|
122
|
-
self,
|
123
|
-
name: str,
|
124
|
-
description: str,
|
125
|
-
embedding: List[float],
|
126
|
-
base_voice_id: Optional[str] = None,
|
127
|
-
language: str = "en",
|
128
|
-
) -> VoiceMetadata:
|
129
|
-
"""Create a new voice.
|
130
|
-
|
131
|
-
Args:
|
132
|
-
name: The name of the voice.
|
133
|
-
description: The description of the voice.
|
134
|
-
embedding: The embedding of the voice. This should be generated with :meth:`clone`.
|
135
|
-
base_voice_id: The ID of the base voice. This should be a valid voice ID if specified.
|
136
|
-
|
137
|
-
Returns:
|
138
|
-
A dictionary containing the voice metadata.
|
139
|
-
"""
|
140
|
-
response = httpx.post(
|
141
|
-
f"{self._http_url()}/voices",
|
142
|
-
headers=self.headers,
|
143
|
-
json={
|
144
|
-
"name": name,
|
145
|
-
"description": description,
|
146
|
-
"embedding": embedding,
|
147
|
-
"base_voice_id": base_voice_id,
|
148
|
-
"language": language,
|
149
|
-
},
|
150
|
-
timeout=self.timeout,
|
151
|
-
)
|
152
|
-
|
153
|
-
if not response.is_success:
|
154
|
-
raise ValueError(f"Failed to create voice. Error: {response.text}")
|
155
|
-
|
156
|
-
return response.json()
|
157
|
-
|
158
|
-
def delete(self, id: str) -> bool:
|
159
|
-
"""Delete a voice by its ID.
|
160
|
-
|
161
|
-
Args:
|
162
|
-
id: The ID of the voice.
|
163
|
-
|
164
|
-
Raises:
|
165
|
-
ValueError: If the request fails.
|
166
|
-
"""
|
167
|
-
response = httpx.delete(
|
168
|
-
f"{self._http_url()}/voices/{id}",
|
169
|
-
headers=self.headers,
|
170
|
-
timeout=self.timeout,
|
171
|
-
)
|
172
|
-
|
173
|
-
if not response.is_success:
|
174
|
-
raise ValueError(f"Failed to delete voice. Error: {response.text}")
|
175
|
-
|
176
|
-
def mix(self, voices: List[Dict[str, Union[str, float]]]) -> List[float]:
|
177
|
-
"""Mix multiple voices together.
|
178
|
-
|
179
|
-
Args:
|
180
|
-
voices: A list of dictionaries, each containing either:
|
181
|
-
- 'id': The ID of an existing voice
|
182
|
-
- 'embedding': A voice embedding
|
183
|
-
AND
|
184
|
-
- 'weight': The weight of the voice in the mix (0.0 to 1.0)
|
185
|
-
|
186
|
-
Returns:
|
187
|
-
The embedding of the mixed voice as a list of floats.
|
188
|
-
|
189
|
-
Raises:
|
190
|
-
ValueError: If the request fails or if the input is invalid.
|
191
|
-
"""
|
192
|
-
url = f"{self._http_url()}/voices/mix"
|
193
|
-
|
194
|
-
if not voices or not isinstance(voices, list):
|
195
|
-
raise ValueError("voices must be a non-empty list")
|
196
|
-
|
197
|
-
response = httpx.post(
|
198
|
-
url,
|
199
|
-
headers=self.headers,
|
200
|
-
json={"voices": voices},
|
201
|
-
timeout=self.timeout,
|
202
|
-
)
|
203
|
-
|
204
|
-
if not response.is_success:
|
205
|
-
raise ValueError(f"Failed to mix voices. Error: {response.text}")
|
206
|
-
|
207
|
-
result = response.json()
|
208
|
-
return result["embedding"]
|