Voice Cloning & Synthesis

Generate high-quality emotional text-to-speech with voice cloning capabilities

POST /api/v1/cloning/synthesize

cURL

curl -X POST https://app.dubformer.ai/api/v1/cloning/synthesize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "text=Hello world, this is a test of our voice cloning technology!" \
  -F "language=en-US" \
  -F "reference_audio_files=@reference_voice_1.wav" \
  -F "reference_audio_files=@reference_voice_2.wav"

import requests

url = 'https://app.dubformer.ai/api/v1/cloning/synthesize'
headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
}

# Prepare the form data
data = {
    'text': 'Hello world, this is a test of our voice cloning technology!',
    'language': 'en-US'
}

# Prepare the files (multiple reference audio files)
files = [
    ('reference_audio_files', open('reference_voice_1.wav', 'rb')),
    ('reference_audio_files', open('reference_voice_2.wav', 'rb'))
]

response = requests.post(url, headers=headers, data=data, files=files)

# Save the audio response
with open('synthesized_audio.wav', 'wb') as f:
    f.write(response.content)

print(f"Audio saved with status: {response.status_code}")

const url = 'https://app.dubformer.ai/api/v1/cloning/synthesize';
const headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
};

// Create form data
const formData = new FormData();
formData.append('text', 'Hello world, this is a test of our voice cloning technology!');
formData.append('language', 'en-US');

// Add reference audio file (assuming you have a file input element)
const fileInput = document.getElementById('audioFile');
if (fileInput.files[0]) {
    formData.append('reference_audio_files', fileInput.files[0]);
}

fetch(url, {
    method: 'POST',
    headers,
    body: formData,
})
.then(response => {
    if (response.ok) {
        return response.blob();
    }
    throw new Error('Network response was not ok');
})
.then(blob => {
    // Create download link for the audio
    const audioUrl = URL.createObjectURL(blob);
    const a = document.createElement('a');
    a.href = audioUrl;
    a.download = 'synthesized_audio.wav';
    a.click();
});

interface ConditioningFeaturesOverride {
    pitch_variance?: 'low' | 'medium' | 'high' | 'auto';
    target_duration?: number | 'auto';
}

interface VoiceCloningRequest {
    text: string;
    language: string;
    reference_audio_files: File[];
}

const url: string = 'https://app.dubformer.ai/api/v1/cloning/synthesize';
const headers: Record<string, string> = {
    'Authorization': 'Bearer YOUR_API_KEY',
};

const formData = new FormData();
const requestData: VoiceCloningRequest = {
    text: 'Hello world, this is a test of our voice cloning technology!',
    language: 'en-US',
    reference_audio_files: []
};

formData.append('text', requestData.text);
formData.append('language', requestData.language);

const fileInput = document.getElementById('audioFile') as HTMLInputElement;
if (fileInput?.files) {
    Array.from(fileInput.files).forEach(file => {
        formData.append('reference_audio_files', file);
    });
}

fetch(url, {
    method: 'POST',
    headers,
    body: formData,
})
.then((response: Response) => {
    if (response.ok) {
        return response.blob();
    }
    throw new Error('Network response was not ok');
})
.then((blob: Blob) => {
    const audioUrl: string = URL.createObjectURL(blob);
    const a = document.createElement('a');
    a.href = audioUrl;
    a.download = 'synthesized_audio.wav';
    a.click();
});

Success Response

# Binary WAV audio file is returned
Content-Type: audio/wav
Content-Length: 524288

# Save to file
curl -X POST https://app.dubformer.ai/api/v1/cloning/synthesize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "text=Hello world!" \
  -F "language=en-US" \
  -F "reference_audio_files=@reference.wav" \
  --output synthesized_audio.wav

{
  "error": "Contact with sales manager for enable API"
}

The Voice Cloning & Synthesis API enables you to generate natural-sounding speech with emotional transfer capabilities. This advanced endpoint uses AI-powered voice cloning technology to synthesize speech that matches the characteristics and emotional tone of reference audio samples.

This endpoint requires a PRO subscription or higher to access voice cloning features.

cURL

curl -X POST https://app.dubformer.ai/api/v1/cloning/synthesize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "text=Hello world, this is a test of our voice cloning technology!" \
  -F "language=en-US" \
  -F "reference_audio_files=@reference_voice_1.wav" \
  -F "reference_audio_files=@reference_voice_2.wav"

import requests

url = 'https://app.dubformer.ai/api/v1/cloning/synthesize'
headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
}

# Prepare the form data
data = {
    'text': 'Hello world, this is a test of our voice cloning technology!',
    'language': 'en-US'
}

# Prepare the files (multiple reference audio files)
files = [
    ('reference_audio_files', open('reference_voice_1.wav', 'rb')),
    ('reference_audio_files', open('reference_voice_2.wav', 'rb'))
]

response = requests.post(url, headers=headers, data=data, files=files)

# Save the audio response
with open('synthesized_audio.wav', 'wb') as f:
    f.write(response.content)

print(f"Audio saved with status: {response.status_code}")

const url = 'https://app.dubformer.ai/api/v1/cloning/synthesize';
const headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
};

// Create form data
const formData = new FormData();
formData.append('text', 'Hello world, this is a test of our voice cloning technology!');
formData.append('language', 'en-US');

// Add reference audio file (assuming you have a file input element)
const fileInput = document.getElementById('audioFile');
if (fileInput.files[0]) {
    formData.append('reference_audio_files', fileInput.files[0]);
}

fetch(url, {
    method: 'POST',
    headers,
    body: formData,
})
.then(response => {
    if (response.ok) {
        return response.blob();
    }
    throw new Error('Network response was not ok');
})
.then(blob => {
    // Create download link for the audio
    const audioUrl = URL.createObjectURL(blob);
    const a = document.createElement('a');
    a.href = audioUrl;
    a.download = 'synthesized_audio.wav';
    a.click();
});

interface ConditioningFeaturesOverride {
    pitch_variance?: 'low' | 'medium' | 'high' | 'auto';
    target_duration?: number | 'auto';
}

interface VoiceCloningRequest {
    text: string;
    language: string;
    reference_audio_files: File[];
}

const url: string = 'https://app.dubformer.ai/api/v1/cloning/synthesize';
const headers: Record<string, string> = {
    'Authorization': 'Bearer YOUR_API_KEY',
};

const formData = new FormData();
const requestData: VoiceCloningRequest = {
    text: 'Hello world, this is a test of our voice cloning technology!',
    language: 'en-US',
    reference_audio_files: []
};

formData.append('text', requestData.text);
formData.append('language', requestData.language);

const fileInput = document.getElementById('audioFile') as HTMLInputElement;
if (fileInput?.files) {
    Array.from(fileInput.files).forEach(file => {
        formData.append('reference_audio_files', file);
    });
}

fetch(url, {
    method: 'POST',
    headers,
    body: formData,
})
.then((response: Response) => {
    if (response.ok) {
        return response.blob();
    }
    throw new Error('Network response was not ok');
})
.then((blob: Blob) => {
    const audioUrl: string = URL.createObjectURL(blob);
    const a = document.createElement('a');
    a.href = audioUrl;
    a.download = 'synthesized_audio.wav';
    a.click();
});

Request Parameters

text · string · required

The text content to be synthesized into speech. Maximum length: 5000 characters.

language · string · required

The target language for speech synthesis with locale (e.g., “en-US”, “fr-FR”, “es-ES”). Must match one of the supported languages from Get Options.

reference_audio_files · file[] · required

One or more audio files containing reference voice samples for cloning.

Supported formats: WAV, MP3, M4A, FLAC
Maximum file size: 50MB per file
Recommended length: 3-30 seconds for optimal results
Quality requirements: Clear audio with minimal background noise

Response

Success Response (200)

audio · binary

Synthesized audio as a binary WAV file.

Content-Type: audio/wav
File Format: 16-bit PCM WAV, 44.1kHz sample rate

Success Response

# Binary WAV audio file is returned
Content-Type: audio/wav
Content-Length: 524288

# Save to file
curl -X POST https://app.dubformer.ai/api/v1/cloning/synthesize \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "text=Hello world!" \
  -F "language=en-US" \
  -F "reference_audio_files=@reference.wav" \
  --output synthesized_audio.wav

{
  "error": "Contact with sales manager for enable API"
}