Create Voice

Create a custom voice from audio samples

POST /voices

cURL

curl -X POST https://api.60db.ai/voices \
  -H "Authorization: Bearer your-api-key" \
  -F "name=My Custom Voice" \
  -F "description=Professional voice for my brand" \
  -F "language=en" \
  -F "gender=female" \
  -F "files=@sample1.mp3" \
  -F "files=@sample2.mp3" \
  -F "files=@sample3.mp3"

import { SixtyDBClient } from "60db";

const client = new SixtyDBClient("your-api-key");

const files = [
  document.querySelector("#file1").files[0],
  document.querySelector("#file2").files[0],
  document.querySelector("#file3").files[0],
];

const voice = await client.createVoice({
  name: "My Custom Voice",
  description: "Professional voice for my brand",
  language: "en",
  gender: "female",
  files: files,
});

console.log("Voice ID:", voice.id);

from sixtydb import SixtyDBClient

client = SixtyDBClient('your-api-key')

files = [
    open('sample1.mp3', 'rb'),
    open('sample2.mp3', 'rb'),
    open('sample3.mp3', 'rb')
]

voice = client.create_voice(
    name='My Custom Voice',
    files=files,
    description='Professional voice for my brand',
    language='en',
    gender='female'
)

print(f"Voice ID: {voice['id']}")

# Close files
for f in files:
    f.close()

Response

{
  "id": "voice-custom-123",
  "name": "My Custom Voice",
  "description": "Professional voice for my brand",
  "language": "en",
  "gender": "female",
  "status": "processing",
  "is_custom": true,
  "created_at": "2026-01-29T11:30:00Z",
  "estimated_completion": "2026-01-29T11:45:00Z"
}

Request

Headers

Authorization · string · required

Bearer token with your API key

Content-Type · string · required

multipart/form-data

Form Data

name · string · required

Name for the custom voice

files · file[] · required

Audio files for voice cloning (minimum 3, maximum 10 files). Each file should be: - Format: MP3, WAV, or FLAC - Duration: 10-60 seconds each - Quality: Clear speech, minimal background noise - Total duration: At least 2 minutes combined

description · string

Description of the voice

language · string

Primary language code (e.g., “en”, “es”)

gender · string

Voice gender: “male”, “female”, or “neutral”

Response

id · string

Unique identifier for the created voice

name · string

Voice name

status · string

Processing status: “processing”, “ready”, “failed”

created_at · string

ISO 8601 timestamp

estimated_completion · string

Estimated time for voice processing to complete

cURL

curl -X POST https://api.60db.ai/voices \
  -H "Authorization: Bearer your-api-key" \
  -F "name=My Custom Voice" \
  -F "description=Professional voice for my brand" \
  -F "language=en" \
  -F "gender=female" \
  -F "files=@sample1.mp3" \
  -F "files=@sample2.mp3" \
  -F "files=@sample3.mp3"

import { SixtyDBClient } from "60db";

const client = new SixtyDBClient("your-api-key");

const files = [
  document.querySelector("#file1").files[0],
  document.querySelector("#file2").files[0],
  document.querySelector("#file3").files[0],
];

const voice = await client.createVoice({
  name: "My Custom Voice",
  description: "Professional voice for my brand",
  language: "en",
  gender: "female",
  files: files,
});

console.log("Voice ID:", voice.id);

from sixtydb import SixtyDBClient

client = SixtyDBClient('your-api-key')

files = [
    open('sample1.mp3', 'rb'),
    open('sample2.mp3', 'rb'),
    open('sample3.mp3', 'rb')
]

voice = client.create_voice(
    name='My Custom Voice',
    files=files,
    description='Professional voice for my brand',
    language='en',
    gender='female'
)

print(f"Voice ID: {voice['id']}")

# Close files
for f in files:
    f.close()

Response

{
  "id": "voice-custom-123",
  "name": "My Custom Voice",
  "description": "Professional voice for my brand",
  "language": "en",
  "gender": "female",
  "status": "processing",
  "is_custom": true,
  "created_at": "2026-01-29T11:30:00Z",
  "estimated_completion": "2026-01-29T11:45:00Z"
}

Voice Cloning Best Practices

Audio Quality

• Use high-quality recordings (at least 44.1kHz sample rate) - Ensure minimal background noise - Avoid music or sound effects - Use consistent recording environment

Content Guidelines

• Include varied sentence structures - Cover different emotions and tones - Include questions, statements, and exclamations - Avoid repetitive content

Technical Requirements

• Minimum 3 files, maximum 10 files - Each file: 10-60 seconds - Total duration: At least 2 minutes - Supported formats: MP3, WAV, FLAC

Voice processing typically takes 10-15 minutes. You’ll receive a webhook notification when your voice is ready to use.