how can I make a voice through the API with more than one file, what needs to be the structure of f

[hearty reef]

AI-generated Answer for neiele

To add a new voice using multiple audio files through the API, you need to structure your request as follows:

1. Headers: Include your API key.
2. Body: Use multipart/form-data to send the voice name and the list of file paths.

Here's an example in Python:

from elevenlabs.client import ElevenLabs

client = ElevenLabs(api_key="YOUR_API_KEY")

client.voices.add(
    name="YourVoiceName",
    files=["path/to/first_audio_file.wav", "path/to/second_audio_file.wav"],
    description="Description of the voice",
    labels="{'label1': 'value1', 'label2': 'value2'}"
)

Request Structure

• Headers:

  • xi-api-key: Your API key.

• Body:

  • name: The name that identifies this voice.
  • files: A list of file paths to audio recordings intended for voice cloning.
  • description (optional): A description of the voice.
  • labels (optional): Serialized labels dictionary for the voice.

Example cURL Request

curl -X POST "https://api.elevenlabs.io/v1/voices/add" \
-H "xi-api-key: YOUR_API_KEY" \
-F "name=YourVoiceName" \
-F "files=@path/to/first_audio_file.wav" \
-F "files=@path/to/second_audio_file.wav" \
-F "description=Description of the voice" \
-F "labels={'label1': 'value1', 'label2': 'value2'}"

This will add a new voice using the provided audio files. Make sure you have the necessary rights or consents to upload and clone the voice samples.

[bright crescent]

what if I store the audio of my files in this structure:
[('files', [('vf_e598a05f-ef8e-44c2-9591-a582c0111a44.wav', <_io.BytesIO object at 0x7fba02f2b830>, 'audio/wav')])]

[hearty reef]

AI-generated Answer for neiele

To create a voice through the API with multiple files, you need to provide a list of file paths to audio recordings. Here is the structure you should follow:

Headers

• xi-api-key: Your API key.

Body (multipart/form-data)

• name (string, required): The name that identifies this voice.
• files (file[], required): A list of file paths to audio recordings intended for voice cloning.
• description (string): How would you describe the voice?
• labels (string): Serialized labels dictionary for the voice.

Example in Python

from elevenlabs.client import ElevenLabs

client = ElevenLabs(
    api_key="YOUR_API_KEY",
)

client.voices.add(
    name="Alex",
    files=[
        ('files', ('vf_e598a05f-ef8e-44c2-9591-a582c0111a44.wav', open('path/to/your/file1.wav', 'rb'), 'audio/wav')),
        ('files', ('another_file.wav', open('path/to/your/file2.wav', 'rb'), 'audio/wav'))
    ],
    description="A description of the voice",
    labels="{}"
)

Make sure each file is correctly formatted and accessible. If you have any issues with account, billing, or other related topics, please open a ticket via the ticket form.

[bright crescent]

I want to use a request, not the python library, and the audio content will be loaded in using BytesIO

[hearty reef]

AI-generated Answer for neiele

To create a voice using the API with multiple audio files using requests and BytesIO, you can structure your request as follows:

import requests
from io import BytesIO

# Your API key
API_KEY = "your_api_key_here"

# URL for the API endpoint
url = "https://api.elevenlabs.io/v1/voice"

# Headers for the request
headers = {
    "xi-api-key": API_KEY
}

# Example audio files loaded into BytesIO
audio_files = [
    ('files', ('file1.wav', BytesIO(b'audio_content_1'), 'audio/wav')),
    ('files', ('file2.wav', BytesIO(b'audio_content_2'), 'audio/wav'))
]

# Data payload for the request
data = {
    "voice_name": "Your Voice Name",
    "description": "Description of the voice"
}

# Make the POST request
response = requests.post(url, headers=headers, data=data, files=audio_files)

# Check the response
if response.ok:
    print("Voice created successfully.")
    print(response.json())
else:
    print("Error:", response.text)

In this example:

• Replace "your_api_key_here" with your actual API key.
• Replace b'audio_content_1' and b'audio_content_2' with the actual binary content of your audio files.
• Adjust the data dictionary to include any additional parameters required by the API.

This will send a multipart/form-data request to the API with multiple audio files.

[bright crescent]

[('files', ('vf_e598a05f-ef8e-44c2-9591-a582c0111a44.wav', <_io.BytesIO object at 0x7f4dc3e73e70>, 'audio/wav')), ('files', ('vf_ca456af8-4d93-4cef-815b-a6200f9e5ffb.wav', <_io.BytesIO object at 0x7f4ddbcc94e0>, 'audio/wav'))]
vf_e598a05f-ef8e-44c2-9591-a582c0111a44.wav
vf_ca456af8-4d93-4cef-815b-a6200f9e5ffb.wav
{'detail': {'status': 'duplicated_files', 'message': 'You cannot upload the same file twice.'}}

I get the duplicated_files error, the files are different