magic_hour 0.40.0__py3-none-any.whl → 0.44.0__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.
- magic_hour/README.md +2 -3
- magic_hour/environment.py +1 -1
- magic_hour/helpers/download.py +2 -0
- magic_hour/resources/v1/README.md +2 -3
- magic_hour/resources/v1/ai_clothes_changer/README.md +13 -14
- magic_hour/resources/v1/ai_face_editor/README.md +26 -27
- magic_hour/resources/v1/ai_gif_generator/README.md +12 -13
- magic_hour/resources/v1/ai_gif_generator/client.py +2 -2
- magic_hour/resources/v1/ai_headshot_generator/README.md +13 -14
- magic_hour/resources/v1/ai_headshot_generator/client.py +2 -2
- magic_hour/resources/v1/ai_image_editor/README.md +24 -17
- magic_hour/resources/v1/ai_image_editor/client.py +40 -10
- magic_hour/resources/v1/ai_image_generator/README.md +26 -18
- magic_hour/resources/v1/ai_image_generator/client.py +14 -6
- magic_hour/resources/v1/ai_image_upscaler/README.md +14 -15
- magic_hour/resources/v1/ai_meme_generator/README.md +12 -13
- magic_hour/resources/v1/ai_photo_editor/README.md +22 -23
- magic_hour/resources/v1/ai_qr_code_generator/README.md +13 -14
- magic_hour/resources/v1/ai_qr_code_generator/client.py +4 -4
- magic_hour/resources/v1/ai_talking_photo/README.md +16 -17
- magic_hour/resources/v1/ai_voice_cloner/README.md +62 -0
- magic_hour/resources/v1/ai_voice_cloner/__init__.py +4 -0
- magic_hour/resources/v1/ai_voice_cloner/client.py +272 -0
- magic_hour/resources/v1/ai_voice_generator/README.md +66 -10
- magic_hour/resources/v1/ai_voice_generator/client.py +122 -0
- magic_hour/resources/v1/animation/README.md +24 -25
- magic_hour/resources/v1/audio_projects/README.md +58 -13
- magic_hour/resources/v1/audio_projects/__init__.py +10 -2
- magic_hour/resources/v1/audio_projects/client.py +137 -0
- magic_hour/resources/v1/audio_projects/client_test.py +520 -0
- magic_hour/resources/v1/auto_subtitle_generator/README.md +15 -16
- magic_hour/resources/v1/client.py +6 -0
- magic_hour/resources/v1/face_detection/README.md +21 -20
- magic_hour/resources/v1/face_swap/README.md +23 -25
- magic_hour/resources/v1/face_swap/client.py +2 -2
- magic_hour/resources/v1/face_swap_photo/README.md +13 -14
- magic_hour/resources/v1/files/README.md +1 -5
- magic_hour/resources/v1/files/upload_urls/README.md +11 -10
- magic_hour/resources/v1/files/upload_urls/client.py +6 -4
- magic_hour/resources/v1/image_background_remover/README.md +11 -12
- magic_hour/resources/v1/image_projects/README.md +12 -16
- magic_hour/resources/v1/image_to_video/README.md +19 -21
- magic_hour/resources/v1/lip_sync/README.md +27 -21
- magic_hour/resources/v1/lip_sync/client.py +15 -0
- magic_hour/resources/v1/photo_colorizer/README.md +10 -11
- magic_hour/resources/v1/text_to_video/README.md +15 -17
- magic_hour/resources/v1/video_projects/README.md +12 -16
- magic_hour/resources/v1/video_to_video/README.md +24 -26
- magic_hour/types/models/__init__.py +2 -0
- magic_hour/types/models/v1_ai_voice_cloner_create_response.py +27 -0
- magic_hour/types/models/v1_audio_projects_get_response.py +1 -1
- magic_hour/types/models/v1_video_projects_get_response.py +1 -1
- magic_hour/types/params/__init__.py +26 -0
- magic_hour/types/params/v1_ai_image_editor_create_body_assets.py +18 -4
- magic_hour/types/params/v1_ai_image_editor_create_body_style.py +13 -0
- magic_hour/types/params/v1_ai_image_editor_generate_body_assets.py +12 -1
- magic_hour/types/params/v1_ai_image_generator_create_body_style.py +16 -0
- magic_hour/types/params/v1_ai_talking_photo_create_body_style.py +6 -4
- magic_hour/types/params/v1_ai_voice_cloner_create_body.py +49 -0
- magic_hour/types/params/v1_ai_voice_cloner_create_body_assets.py +33 -0
- magic_hour/types/params/v1_ai_voice_cloner_create_body_style.py +28 -0
- magic_hour/types/params/v1_ai_voice_cloner_generate_body_assets.py +28 -0
- magic_hour/types/params/v1_ai_voice_generator_create_body_style.py +382 -2
- magic_hour/types/params/v1_face_swap_create_body_style.py +1 -1
- magic_hour/types/params/v1_files_upload_urls_create_body_items_item.py +1 -1
- magic_hour/types/params/v1_lip_sync_create_body.py +12 -0
- magic_hour/types/params/v1_lip_sync_create_body_style.py +37 -0
- magic_hour/types/params/v1_video_to_video_create_body.py +1 -1
- magic_hour/types/params/v1_video_to_video_create_body_style.py +32 -4
- {magic_hour-0.40.0.dist-info → magic_hour-0.44.0.dist-info}/METADATA +77 -62
- {magic_hour-0.40.0.dist-info → magic_hour-0.44.0.dist-info}/RECORD +73 -63
- {magic_hour-0.40.0.dist-info → magic_hour-0.44.0.dist-info}/LICENSE +0 -0
- {magic_hour-0.40.0.dist-info → magic_hour-0.44.0.dist-info}/WHEEL +0 -0
|
@@ -2,8 +2,6 @@
|
|
|
2
2
|
|
|
3
3
|
## Module Functions
|
|
4
4
|
|
|
5
|
-
|
|
6
|
-
|
|
7
5
|
<!-- CUSTOM DOCS START -->
|
|
8
6
|
|
|
9
7
|
### Ai Talking Photo Generate Workflow <a name="generate"></a>
|
|
@@ -69,6 +67,7 @@ res = await client.v1.ai_talking_photo.generate(
|
|
|
69
67
|
```
|
|
70
68
|
|
|
71
69
|
<!-- CUSTOM DOCS END -->
|
|
70
|
+
|
|
72
71
|
### AI Talking Photo <a name="create"></a>
|
|
73
72
|
|
|
74
73
|
Create a talking photo from an image and audio or text input.
|
|
@@ -77,17 +76,17 @@ Create a talking photo from an image and audio or text input.
|
|
|
77
76
|
|
|
78
77
|
#### Parameters
|
|
79
78
|
|
|
80
|
-
| Parameter
|
|
81
|
-
|
|
82
|
-
| `assets`
|
|
83
|
-
| `└─ audio_file_path` |
|
|
84
|
-
| `└─ image_file_path` |
|
|
85
|
-
| `end_seconds`
|
|
86
|
-
| `start_seconds`
|
|
87
|
-
| `name`
|
|
88
|
-
| `style`
|
|
89
|
-
| `└─ generation_mode` |
|
|
90
|
-
| `└─ intensity`
|
|
79
|
+
| Parameter | Required | Description | Example |
|
|
80
|
+
| -------------------- | :------: | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- |
|
|
81
|
+
| `assets` | ✓ | Provide the assets for creating a talking photo | `{"audio_file_path": "api-assets/id/1234.mp3", "image_file_path": "api-assets/id/1234.png"}` |
|
|
82
|
+
| `└─ audio_file_path` | ✓ | The audio file to sync with the image. This value is either - a direct URL to the video file - `file_path` field from the response of the [upload urls API](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls). Please refer to the [Input File documentation](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls#input-file) to learn more. | `"api-assets/id/1234.mp3"` |
|
|
83
|
+
| `└─ image_file_path` | ✓ | The source image to animate. This value is either - a direct URL to the video file - `file_path` field from the response of the [upload urls API](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls). Please refer to the [Input File documentation](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls#input-file) to learn more. | `"api-assets/id/1234.png"` |
|
|
84
|
+
| `end_seconds` | ✓ | The end time of the input audio in seconds. The maximum duration allowed is 60 seconds. | `15.0` |
|
|
85
|
+
| `start_seconds` | ✓ | The start time of the input audio in seconds. The maximum duration allowed is 60 seconds. | `0.0` |
|
|
86
|
+
| `name` | ✗ | The name of image. This value is mainly used for your own identification of the image. | `"Talking Photo image"` |
|
|
87
|
+
| `style` | ✗ | Attributes used to dictate the style of the output | `{"generation_mode": "pro", "intensity": 1.5}` |
|
|
88
|
+
| `└─ generation_mode` | ✗ | Controls overall motion style. * `pro` - Higher fidelity, realistic detail, accurate lip sync, and faster generation. * `standard` - More expressive motion, but lower visual fidelity. * `expressive` - More motion and facial expressiveness; may introduce visual artifacts. (Deprecated: passing this value will be treated as `standard`) * `stable` - Reduced motion for cleaner output; may result in minimal animation. (Deprecated: passing this value will be treated as `pro`) | `"pro"` |
|
|
89
|
+
| `└─ intensity` | ✗ | Note: this value is only applicable when generation_mode is `expressive`. The value can include up to 2 decimal places. * Lower values yield more stability but can suppress mouth movement. * Higher values increase motion and expressiveness, with a higher risk of distortion. | `1.5` |
|
|
91
90
|
|
|
92
91
|
#### Synchronous Client
|
|
93
92
|
|
|
@@ -105,7 +104,6 @@ res = client.v1.ai_talking_photo.create(
|
|
|
105
104
|
start_seconds=0.0,
|
|
106
105
|
name="Talking Photo image",
|
|
107
106
|
)
|
|
108
|
-
|
|
109
107
|
```
|
|
110
108
|
|
|
111
109
|
#### Asynchronous Client
|
|
@@ -124,15 +122,16 @@ res = await client.v1.ai_talking_photo.create(
|
|
|
124
122
|
start_seconds=0.0,
|
|
125
123
|
name="Talking Photo image",
|
|
126
124
|
)
|
|
127
|
-
|
|
128
125
|
```
|
|
129
126
|
|
|
130
127
|
#### Response
|
|
131
128
|
|
|
132
129
|
##### Type
|
|
130
|
+
|
|
133
131
|
[V1AiTalkingPhotoCreateResponse](/magic_hour/types/models/v1_ai_talking_photo_create_response.py)
|
|
134
132
|
|
|
135
133
|
##### Example
|
|
136
|
-
`{"credits_charged": 450, "estimated_frame_cost": 450, "id": "cuid-example"}`
|
|
137
|
-
|
|
138
134
|
|
|
135
|
+
```python
|
|
136
|
+
{"credits_charged": 450, "estimated_frame_cost": 450, "id": "cuid-example"}
|
|
137
|
+
```
|
|
@@ -0,0 +1,62 @@
|
|
|
1
|
+
# v1.ai_voice_cloner
|
|
2
|
+
|
|
3
|
+
## Module Functions
|
|
4
|
+
|
|
5
|
+
### AI Voice Cloner <a name="create"></a>
|
|
6
|
+
|
|
7
|
+
Clone a voice from an audio sample and generate speech.
|
|
8
|
+
|
|
9
|
+
- Each character costs 0.05 credits.
|
|
10
|
+
- The cost is rounded up to the nearest whole number
|
|
11
|
+
|
|
12
|
+
**API Endpoint**: `POST /v1/ai-voice-cloner`
|
|
13
|
+
|
|
14
|
+
#### Parameters
|
|
15
|
+
|
|
16
|
+
| Parameter | Required | Description | Example |
|
|
17
|
+
| -------------------- | :------: | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------- |
|
|
18
|
+
| `assets` | ✓ | Provide the assets for voice cloning. | `{"audio_file_path": "api-assets/id/1234.mp3"}` |
|
|
19
|
+
| `└─ audio_file_path` | ✓ | The audio used to clone the voice. This value is either - a direct URL to the video file - `file_path` field from the response of the [upload urls API](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls). Please refer to the [Input File documentation](https://docs.magichour.ai/api-reference/files/generate-asset-upload-urls#input-file) to learn more. | `"api-assets/id/1234.mp3"` |
|
|
20
|
+
| `style` | ✓ | | `{"prompt": "Hello, this is my cloned voice."}` |
|
|
21
|
+
| `└─ prompt` | ✓ | Text used to generate speech from the cloned voice. The character limit is 1000 characters. | `"Hello, this is my cloned voice."` |
|
|
22
|
+
| `name` | ✗ | The name of audio. This value is mainly used for your own identification of the audio. | `"Voice Cloner audio"` |
|
|
23
|
+
|
|
24
|
+
#### Synchronous Client
|
|
25
|
+
|
|
26
|
+
```python
|
|
27
|
+
from magic_hour import Client
|
|
28
|
+
from os import getenv
|
|
29
|
+
|
|
30
|
+
client = Client(token=getenv("API_TOKEN"))
|
|
31
|
+
res = client.v1.ai_voice_cloner.create(
|
|
32
|
+
assets={"audio_file_path": "api-assets/id/1234.mp3"},
|
|
33
|
+
style={"prompt": "Hello, this is my cloned voice."},
|
|
34
|
+
name="Voice Cloner audio",
|
|
35
|
+
)
|
|
36
|
+
```
|
|
37
|
+
|
|
38
|
+
#### Asynchronous Client
|
|
39
|
+
|
|
40
|
+
```python
|
|
41
|
+
from magic_hour import AsyncClient
|
|
42
|
+
from os import getenv
|
|
43
|
+
|
|
44
|
+
client = AsyncClient(token=getenv("API_TOKEN"))
|
|
45
|
+
res = await client.v1.ai_voice_cloner.create(
|
|
46
|
+
assets={"audio_file_path": "api-assets/id/1234.mp3"},
|
|
47
|
+
style={"prompt": "Hello, this is my cloned voice."},
|
|
48
|
+
name="Voice Cloner audio",
|
|
49
|
+
)
|
|
50
|
+
```
|
|
51
|
+
|
|
52
|
+
#### Response
|
|
53
|
+
|
|
54
|
+
##### Type
|
|
55
|
+
|
|
56
|
+
[V1AiVoiceClonerCreateResponse](/magic_hour/types/models/v1_ai_voice_cloner_create_response.py)
|
|
57
|
+
|
|
58
|
+
##### Example
|
|
59
|
+
|
|
60
|
+
```python
|
|
61
|
+
{"credits_charged": 1, "id": "cuid-example"}
|
|
62
|
+
```
|
|
@@ -0,0 +1,272 @@
|
|
|
1
|
+
import typing
|
|
2
|
+
|
|
3
|
+
from magic_hour.helpers.logger import get_sdk_logger
|
|
4
|
+
from magic_hour.resources.v1.audio_projects.client import (
|
|
5
|
+
AsyncAudioProjectsClient,
|
|
6
|
+
AudioProjectsClient,
|
|
7
|
+
)
|
|
8
|
+
from magic_hour.resources.v1.files.client import AsyncFilesClient, FilesClient
|
|
9
|
+
from magic_hour.types import models, params
|
|
10
|
+
from make_api_request import (
|
|
11
|
+
AsyncBaseClient,
|
|
12
|
+
RequestOptions,
|
|
13
|
+
SyncBaseClient,
|
|
14
|
+
default_request_options,
|
|
15
|
+
to_encodable,
|
|
16
|
+
type_utils,
|
|
17
|
+
)
|
|
18
|
+
|
|
19
|
+
|
|
20
|
+
logger = get_sdk_logger(__name__)
|
|
21
|
+
|
|
22
|
+
|
|
23
|
+
class AiVoiceClonerClient:
|
|
24
|
+
def __init__(self, *, base_client: SyncBaseClient):
|
|
25
|
+
self._base_client = base_client
|
|
26
|
+
|
|
27
|
+
def generate(
|
|
28
|
+
self,
|
|
29
|
+
*,
|
|
30
|
+
assets: params.V1AiVoiceClonerGenerateBodyAssets,
|
|
31
|
+
style: params.V1AiVoiceClonerCreateBodyStyle,
|
|
32
|
+
name: typing.Union[
|
|
33
|
+
typing.Optional[str], type_utils.NotGiven
|
|
34
|
+
] = type_utils.NOT_GIVEN,
|
|
35
|
+
wait_for_completion: bool = True,
|
|
36
|
+
download_outputs: bool = True,
|
|
37
|
+
download_directory: typing.Optional[str] = None,
|
|
38
|
+
request_options: typing.Optional[RequestOptions] = None,
|
|
39
|
+
):
|
|
40
|
+
"""
|
|
41
|
+
Generate cloned voice audio (alias for create with additional functionality).
|
|
42
|
+
|
|
43
|
+
Clone a voice from an audio sample and generate speech with automatic file upload,
|
|
44
|
+
completion waiting, and output downloading.
|
|
45
|
+
|
|
46
|
+
Args:
|
|
47
|
+
assets: Provide the assets for voice cloning (local file paths will be auto-uploaded)
|
|
48
|
+
style: Voice cloning style parameters including the text prompt
|
|
49
|
+
name: The name of audio. This value is mainly used for your own identification of the audio.
|
|
50
|
+
wait_for_completion: Whether to wait for the audio project to complete
|
|
51
|
+
download_outputs: Whether to download the outputs
|
|
52
|
+
download_directory: The directory to download the outputs to. If not provided, the outputs will be downloaded to the current working directory
|
|
53
|
+
request_options: Additional options to customize the HTTP request
|
|
54
|
+
|
|
55
|
+
Returns:
|
|
56
|
+
V1AudioProjectsGetResponseWithDownloads: The response from the AI Voice Cloner API with the downloaded paths if `download_outputs` is True.
|
|
57
|
+
|
|
58
|
+
Examples:
|
|
59
|
+
```py
|
|
60
|
+
response = client.v1.ai_voice_cloner.generate(
|
|
61
|
+
assets={"audio_file_path": "path/to/voice_sample.mp3"},
|
|
62
|
+
style={"prompt": "Hello, this is my cloned voice speaking."},
|
|
63
|
+
name="Cloned Voice Audio",
|
|
64
|
+
wait_for_completion=True,
|
|
65
|
+
download_outputs=True,
|
|
66
|
+
download_directory="outputs/",
|
|
67
|
+
)
|
|
68
|
+
```
|
|
69
|
+
"""
|
|
70
|
+
|
|
71
|
+
file_client = FilesClient(base_client=self._base_client)
|
|
72
|
+
|
|
73
|
+
audio_file_path = assets["audio_file_path"]
|
|
74
|
+
assets["audio_file_path"] = file_client.upload_file(file=audio_file_path)
|
|
75
|
+
|
|
76
|
+
create_response = self.create(
|
|
77
|
+
assets=assets,
|
|
78
|
+
style=style,
|
|
79
|
+
name=name,
|
|
80
|
+
request_options=request_options,
|
|
81
|
+
)
|
|
82
|
+
logger.info(f"AI Voice Cloner response: {create_response}")
|
|
83
|
+
|
|
84
|
+
audio_projects_client = AudioProjectsClient(base_client=self._base_client)
|
|
85
|
+
response = audio_projects_client.check_result(
|
|
86
|
+
id=create_response.id,
|
|
87
|
+
wait_for_completion=wait_for_completion,
|
|
88
|
+
download_outputs=download_outputs,
|
|
89
|
+
download_directory=download_directory,
|
|
90
|
+
)
|
|
91
|
+
|
|
92
|
+
return response
|
|
93
|
+
|
|
94
|
+
def create(
|
|
95
|
+
self,
|
|
96
|
+
*,
|
|
97
|
+
assets: params.V1AiVoiceClonerCreateBodyAssets,
|
|
98
|
+
style: params.V1AiVoiceClonerCreateBodyStyle,
|
|
99
|
+
name: typing.Union[
|
|
100
|
+
typing.Optional[str], type_utils.NotGiven
|
|
101
|
+
] = type_utils.NOT_GIVEN,
|
|
102
|
+
request_options: typing.Optional[RequestOptions] = None,
|
|
103
|
+
) -> models.V1AiVoiceClonerCreateResponse:
|
|
104
|
+
"""
|
|
105
|
+
AI Voice Cloner
|
|
106
|
+
|
|
107
|
+
Clone a voice from an audio sample and generate speech.
|
|
108
|
+
* Each character costs 0.05 credits.
|
|
109
|
+
* The cost is rounded up to the nearest whole number
|
|
110
|
+
|
|
111
|
+
POST /v1/ai-voice-cloner
|
|
112
|
+
|
|
113
|
+
Args:
|
|
114
|
+
name: The name of audio. This value is mainly used for your own identification of the audio.
|
|
115
|
+
assets: Provide the assets for voice cloning.
|
|
116
|
+
style: V1AiVoiceClonerCreateBodyStyle
|
|
117
|
+
request_options: Additional options to customize the HTTP request
|
|
118
|
+
|
|
119
|
+
Returns:
|
|
120
|
+
Success
|
|
121
|
+
|
|
122
|
+
Raises:
|
|
123
|
+
ApiError: A custom exception class that provides additional context
|
|
124
|
+
for API errors, including the HTTP status code and response body.
|
|
125
|
+
|
|
126
|
+
Examples:
|
|
127
|
+
```py
|
|
128
|
+
client.v1.ai_voice_cloner.create(
|
|
129
|
+
assets={"audio_file_path": "api-assets/id/1234.mp3"},
|
|
130
|
+
style={"prompt": "Hello, this is my cloned voice."},
|
|
131
|
+
name="Voice Cloner audio",
|
|
132
|
+
)
|
|
133
|
+
```
|
|
134
|
+
"""
|
|
135
|
+
_json = to_encodable(
|
|
136
|
+
item={"name": name, "assets": assets, "style": style},
|
|
137
|
+
dump_with=params._SerializerV1AiVoiceClonerCreateBody,
|
|
138
|
+
)
|
|
139
|
+
return self._base_client.request(
|
|
140
|
+
method="POST",
|
|
141
|
+
path="/v1/ai-voice-cloner",
|
|
142
|
+
auth_names=["bearerAuth"],
|
|
143
|
+
json=_json,
|
|
144
|
+
cast_to=models.V1AiVoiceClonerCreateResponse,
|
|
145
|
+
request_options=request_options or default_request_options(),
|
|
146
|
+
)
|
|
147
|
+
|
|
148
|
+
|
|
149
|
+
class AsyncAiVoiceClonerClient:
|
|
150
|
+
def __init__(self, *, base_client: AsyncBaseClient):
|
|
151
|
+
self._base_client = base_client
|
|
152
|
+
|
|
153
|
+
async def generate(
|
|
154
|
+
self,
|
|
155
|
+
*,
|
|
156
|
+
assets: params.V1AiVoiceClonerGenerateBodyAssets,
|
|
157
|
+
style: params.V1AiVoiceClonerCreateBodyStyle,
|
|
158
|
+
name: typing.Union[
|
|
159
|
+
typing.Optional[str], type_utils.NotGiven
|
|
160
|
+
] = type_utils.NOT_GIVEN,
|
|
161
|
+
wait_for_completion: bool = True,
|
|
162
|
+
download_outputs: bool = True,
|
|
163
|
+
download_directory: typing.Optional[str] = None,
|
|
164
|
+
request_options: typing.Optional[RequestOptions] = None,
|
|
165
|
+
):
|
|
166
|
+
"""
|
|
167
|
+
Generate cloned voice audio (alias for create with additional functionality).
|
|
168
|
+
|
|
169
|
+
Clone a voice from an audio sample and generate speech with automatic file upload,
|
|
170
|
+
completion waiting, and output downloading.
|
|
171
|
+
|
|
172
|
+
Args:
|
|
173
|
+
assets: Provide the assets for voice cloning (local file paths will be auto-uploaded)
|
|
174
|
+
style: Voice cloning style parameters including the text prompt
|
|
175
|
+
name: The name of audio. This value is mainly used for your own identification of the audio.
|
|
176
|
+
wait_for_completion: Whether to wait for the audio project to complete
|
|
177
|
+
download_outputs: Whether to download the outputs
|
|
178
|
+
download_directory: The directory to download the outputs to. If not provided, the outputs will be downloaded to the current working directory
|
|
179
|
+
request_options: Additional options to customize the HTTP request
|
|
180
|
+
|
|
181
|
+
Returns:
|
|
182
|
+
V1AudioProjectsGetResponseWithDownloads: The response from the AI Voice Cloner API with the downloaded paths if `download_outputs` is True.
|
|
183
|
+
|
|
184
|
+
Examples:
|
|
185
|
+
```py
|
|
186
|
+
response = await client.v1.ai_voice_cloner.generate(
|
|
187
|
+
assets={"audio_file_path": "path/to/voice_sample.mp3"},
|
|
188
|
+
style={"prompt": "Hello, this is my cloned voice speaking."},
|
|
189
|
+
name="Cloned Voice Audio",
|
|
190
|
+
wait_for_completion=True,
|
|
191
|
+
download_outputs=True,
|
|
192
|
+
download_directory="outputs/",
|
|
193
|
+
)
|
|
194
|
+
```
|
|
195
|
+
"""
|
|
196
|
+
|
|
197
|
+
file_client = AsyncFilesClient(base_client=self._base_client)
|
|
198
|
+
|
|
199
|
+
audio_file_path = assets["audio_file_path"]
|
|
200
|
+
assets["audio_file_path"] = await file_client.upload_file(file=audio_file_path)
|
|
201
|
+
|
|
202
|
+
create_response = await self.create(
|
|
203
|
+
assets=assets,
|
|
204
|
+
style=style,
|
|
205
|
+
name=name,
|
|
206
|
+
request_options=request_options,
|
|
207
|
+
)
|
|
208
|
+
logger.info(f"AI Voice Cloner response: {create_response}")
|
|
209
|
+
|
|
210
|
+
audio_projects_client = AsyncAudioProjectsClient(base_client=self._base_client)
|
|
211
|
+
response = await audio_projects_client.check_result(
|
|
212
|
+
id=create_response.id,
|
|
213
|
+
wait_for_completion=wait_for_completion,
|
|
214
|
+
download_outputs=download_outputs,
|
|
215
|
+
download_directory=download_directory,
|
|
216
|
+
)
|
|
217
|
+
|
|
218
|
+
return response
|
|
219
|
+
|
|
220
|
+
async def create(
|
|
221
|
+
self,
|
|
222
|
+
*,
|
|
223
|
+
assets: params.V1AiVoiceClonerCreateBodyAssets,
|
|
224
|
+
style: params.V1AiVoiceClonerCreateBodyStyle,
|
|
225
|
+
name: typing.Union[
|
|
226
|
+
typing.Optional[str], type_utils.NotGiven
|
|
227
|
+
] = type_utils.NOT_GIVEN,
|
|
228
|
+
request_options: typing.Optional[RequestOptions] = None,
|
|
229
|
+
) -> models.V1AiVoiceClonerCreateResponse:
|
|
230
|
+
"""
|
|
231
|
+
AI Voice Cloner
|
|
232
|
+
|
|
233
|
+
Clone a voice from an audio sample and generate speech.
|
|
234
|
+
* Each character costs 0.05 credits.
|
|
235
|
+
* The cost is rounded up to the nearest whole number
|
|
236
|
+
|
|
237
|
+
POST /v1/ai-voice-cloner
|
|
238
|
+
|
|
239
|
+
Args:
|
|
240
|
+
name: The name of audio. This value is mainly used for your own identification of the audio.
|
|
241
|
+
assets: Provide the assets for voice cloning.
|
|
242
|
+
style: V1AiVoiceClonerCreateBodyStyle
|
|
243
|
+
request_options: Additional options to customize the HTTP request
|
|
244
|
+
|
|
245
|
+
Returns:
|
|
246
|
+
Success
|
|
247
|
+
|
|
248
|
+
Raises:
|
|
249
|
+
ApiError: A custom exception class that provides additional context
|
|
250
|
+
for API errors, including the HTTP status code and response body.
|
|
251
|
+
|
|
252
|
+
Examples:
|
|
253
|
+
```py
|
|
254
|
+
await client.v1.ai_voice_cloner.create(
|
|
255
|
+
assets={"audio_file_path": "api-assets/id/1234.mp3"},
|
|
256
|
+
style={"prompt": "Hello, this is my cloned voice."},
|
|
257
|
+
name="Voice Cloner audio",
|
|
258
|
+
)
|
|
259
|
+
```
|
|
260
|
+
"""
|
|
261
|
+
_json = to_encodable(
|
|
262
|
+
item={"name": name, "assets": assets, "style": style},
|
|
263
|
+
dump_with=params._SerializerV1AiVoiceClonerCreateBody,
|
|
264
|
+
)
|
|
265
|
+
return await self._base_client.request(
|
|
266
|
+
method="POST",
|
|
267
|
+
path="/v1/ai-voice-cloner",
|
|
268
|
+
auth_names=["bearerAuth"],
|
|
269
|
+
json=_json,
|
|
270
|
+
cast_to=models.V1AiVoiceClonerCreateResponse,
|
|
271
|
+
request_options=request_options or default_request_options(),
|
|
272
|
+
)
|
|
@@ -2,6 +2,62 @@
|
|
|
2
2
|
|
|
3
3
|
## Module Functions
|
|
4
4
|
|
|
5
|
+
<!-- CUSTOM DOCS START -->
|
|
6
|
+
|
|
7
|
+
### Ai Talking Photo Generate Workflow <a name="generate"></a>
|
|
8
|
+
|
|
9
|
+
The workflow performs the following action
|
|
10
|
+
|
|
11
|
+
1. upload local assets to Magic Hour storage. So you can pass in a local path instead of having to upload files yourself
|
|
12
|
+
2. trigger a generation
|
|
13
|
+
3. poll for a completion status. This is configurable
|
|
14
|
+
4. if success, download the output to local directory
|
|
15
|
+
|
|
16
|
+
> [!TIP]
|
|
17
|
+
> This is the recommended way to use the SDK unless you have specific needs where it is necessary to split up the actions.
|
|
18
|
+
|
|
19
|
+
#### Parameters
|
|
20
|
+
|
|
21
|
+
In Additional to the parameters listed in the `.create` section below, `.generate` introduces 3 new parameters:
|
|
22
|
+
|
|
23
|
+
- `wait_for_completion` (bool, default True): Whether to wait for the project to complete.
|
|
24
|
+
- `download_outputs` (bool, default True): Whether to download the generated files
|
|
25
|
+
- `download_directory` (str, optional): Directory to save downloaded files (defaults to current directory)
|
|
26
|
+
|
|
27
|
+
#### Synchronous Client
|
|
28
|
+
|
|
29
|
+
```python
|
|
30
|
+
from magic_hour import Client
|
|
31
|
+
from os import getenv
|
|
32
|
+
|
|
33
|
+
client = Client(token=getenv("API_TOKEN"))
|
|
34
|
+
res = client.v1.ai_voice_generator.generate(
|
|
35
|
+
style={"prompt": "Hello, how are you?", "voice_name": "Elon Musk"},
|
|
36
|
+
name="Voice Generator audio",
|
|
37
|
+
wait_for_completion=True,
|
|
38
|
+
download_outputs=True,
|
|
39
|
+
download_directory="outputs"
|
|
40
|
+
)
|
|
41
|
+
```
|
|
42
|
+
|
|
43
|
+
#### Asynchronous Client
|
|
44
|
+
|
|
45
|
+
```python
|
|
46
|
+
from magic_hour import AsyncClient
|
|
47
|
+
from os import getenv
|
|
48
|
+
|
|
49
|
+
client = AsyncClient(token=getenv("API_TOKEN"))
|
|
50
|
+
res = await client.v1.ai_voice_generator.generate(
|
|
51
|
+
style={"prompt": "Hello, how are you?", "voice_name": "Elon Musk"},
|
|
52
|
+
name="Voice Generator audio",
|
|
53
|
+
wait_for_completion=True,
|
|
54
|
+
download_outputs=True,
|
|
55
|
+
download_directory="outputs"
|
|
56
|
+
)
|
|
57
|
+
```
|
|
58
|
+
|
|
59
|
+
<!-- CUSTOM DOCS END -->
|
|
60
|
+
|
|
5
61
|
### AI Voice Generator <a name="create"></a>
|
|
6
62
|
|
|
7
63
|
Generate speech from text. Each character costs 0.05 credits. The cost is rounded up to the nearest whole number.
|
|
@@ -10,12 +66,12 @@ Generate speech from text. Each character costs 0.05 credits. The cost is rounde
|
|
|
10
66
|
|
|
11
67
|
#### Parameters
|
|
12
68
|
|
|
13
|
-
| Parameter
|
|
14
|
-
|
|
15
|
-
| `style`
|
|
16
|
-
| `└─ prompt`
|
|
17
|
-
| `└─ voice_name` |
|
|
18
|
-
| `name`
|
|
69
|
+
| Parameter | Required | Description | Example |
|
|
70
|
+
| --------------- | :------: | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- |
|
|
71
|
+
| `style` | ✓ | The content used to generate speech. | `{"prompt": "Hello, how are you?", "voice_name": "Elon Musk"}` |
|
|
72
|
+
| `└─ prompt` | ✓ | Text used to generate speech. Starter tier users can use up to 1000 characters, while Creator, Pro, or Business users can use up to 1000. | `"Hello, how are you?"` |
|
|
73
|
+
| `└─ voice_name` | ✓ | The voice to use for the speech. Available voices: Elon Musk, Mark Zuckerberg, Joe Rogan, Barack Obama, Morgan Freeman, Kanye West, Donald Trump, Joe Biden, Kim Kardashian, Taylor Swift, James Earl Jones, Samuel L. Jackson, Jeff Goldblum, David Attenborough, Sean Connery, Cillian Murphy, Anne Hathaway, Julia Roberts, Natalie Portman, Steve Carell, Amy Poehler, Stephen Colbert, Jimmy Fallon, David Letterman, Alex Trebek, Katy Perry, Prince, Kevin Bacon, Tom Hiddleston, Adam Driver, Alan Rickman, Alexz Johnson, Ana Gasteyer, Andrew Rannells, Arden Cho, Bear Grylls, Ben McKenzie, Ben Stiller, Ben Whishaw, Billie Joe Armstrong, Bingbing Li, Bob Barker, Booboo Stewart, Bradley Steven Perry, Bruno Mars, Caity Lotz, Cameron Boyce, Candice Accola, Carrie Underwood, Casey Affleck, Caterina Scorsone, Cedric the Entertainer, Chace Crawford, Chadwick Boseman, Charlie Day, Chris Hemsworth, Chris Martin, Christopher Mintz-Plasse, Dan Fogler, Dan Stevens, Daniel Dae Kim, Danielle Panabaker, Dave Bautista, David Schwimmer, Denis Leary, Derek Mears, Diego Luna, Donald Glover, Donnie Yen, Doutzen Kroes, Dove Cameron, Dr. Dre, Drake Bell, Elle Fanning, Ernie Hudson, Fergie, Forest Whitaker, Francia Raisa, Freddie Highmore, Gillian Jacobs, Gina Carano, Ginnifer Goodwin, Gordon Ramsay, Guy Pearce, Gwendoline Christie, Hailee Steinfeld, Howie Mandel, Hugh Jackman, Hugh Laurie, J. K. Simmons, Jack Black, Jared Leto, Jennifer Carpenter, Kesha, Kris Jenner, Kristen Bell, Lorde, Matt Smith, Marilyn Monroe, Charlie Chaplin, Albert Einstein, Abraham Lincoln, John F. Kennedy, Lucille Ball, A.R. Rahman, Aamir Khan, Ajay Devgn, Akshay Kumar, Alain Delon, Alan Alda, Alan Cumming, Amitabh Bachchan, Ang Lee, Ansel Elgort, Anthony Anderson, Anthony Mackie, Armie Hammer, Asa Butterfield, B.J. Novak, Barbara Eden, Betty White, Bill Nighy, Bill Pullman, Blake Shelton, Bonnie Wright, Brad Paisley, Brendan Gleeson, Brian Cox, Bruno Ganz, Burt Reynolds, Carrie Fisher, Charles Dance, Chiwetel Ejiofor, Chris Pine, Christina Hendricks, Christina Ricci, Cyndi Lauper, Dakota Fanning, Damian Lewis, Dan Aykroyd, Daniel Craig, David Oyelowo, David Tennant, Diane Keaton, Diane Kruger, Dick Van Dyke, Domhnall Gleeson, Dominic Cooper, Donald Sutherland, Drew Carey, Eartha Kitt, Eddie Izzard, Edward Asner, Eli Roth, Elisabeth Moss, Ellen Burstyn, Emile Hirsch, Ezra Miller, Felicity Jones, Fiona Shaw, Florence Henderson, Freida Pinto, Geena Davis, Gemma Arterton, Geri Halliwell, Glenn Close, Gloria Steinem, Greta Gerwig, Gugu Mbatha-Raw, Hans Zimmer, Harry Connick Jr., Harvey Keitel, Helena Bonham Carter, Henry Cavill, Hilary Swank, Hugh Bonneville, Idina Menzel, Imelda Staunton, Ingrid Bergman, Irrfan Khan, Isla Fisher, Iwan Rheon, Jack Lemmon, Janet Jackson, Jason Bateman, Jason Segel, Jennifer Coolidge, Johnny Galecki, Jon Favreau, Joseph Gordon-Levitt, Josh Brolin, Josh Gad, Josh Groban, Julia Louis-Dreyfus, Kristen Stewart, Kristen Wiig, Rooney Mara, Caitriona Balfe, J.J. Abrams, Zoe Saldana | `"Elon Musk"` |
|
|
74
|
+
| `name` | ✗ | The name of audio. This value is mainly used for your own identification of the audio. | `"Voice Generator audio"` |
|
|
19
75
|
|
|
20
76
|
#### Synchronous Client
|
|
21
77
|
|
|
@@ -28,7 +84,6 @@ res = client.v1.ai_voice_generator.create(
|
|
|
28
84
|
style={"prompt": "Hello, how are you?", "voice_name": "Elon Musk"},
|
|
29
85
|
name="Voice Generator audio",
|
|
30
86
|
)
|
|
31
|
-
|
|
32
87
|
```
|
|
33
88
|
|
|
34
89
|
#### Asynchronous Client
|
|
@@ -42,15 +97,16 @@ res = await client.v1.ai_voice_generator.create(
|
|
|
42
97
|
style={"prompt": "Hello, how are you?", "voice_name": "Elon Musk"},
|
|
43
98
|
name="Voice Generator audio",
|
|
44
99
|
)
|
|
45
|
-
|
|
46
100
|
```
|
|
47
101
|
|
|
48
102
|
#### Response
|
|
49
103
|
|
|
50
104
|
##### Type
|
|
105
|
+
|
|
51
106
|
[V1AiVoiceGeneratorCreateResponse](/magic_hour/types/models/v1_ai_voice_generator_create_response.py)
|
|
52
107
|
|
|
53
108
|
##### Example
|
|
54
|
-
`{"credits_charged": 1, "id": "cuid-example"}`
|
|
55
|
-
|
|
56
109
|
|
|
110
|
+
```python
|
|
111
|
+
{"credits_charged": 1, "id": "cuid-example"}
|
|
112
|
+
```
|