Skip to main content
POST
/
api
/
v2
/
voice-clones
Create Voice Clone
curl --request POST \
  --url https://api.buena.ai/api/v2/voice-clones \
  --header 'Authorization: <authorization>' \
  --header 'Content-Type: <content-type>' \
  --data '
{
  "name": "<string>",
  "description": "<string>",
  "files": [
    null
  ],
  "settings": {
    "stability": 123,
    "clarity": 123,
    "use_speaker_boost": true
  }
}
'
{
  "success": true,
  "data": {
    "voiceId": "pNInz6obpgDQGcFmaJgB",
    "name": "Professional Voice",
    "description": "Professional outreach voice for LinkedIn",
    "sampleCount": 3,
    "createdAt": "2024-01-20T15:30:00.000Z"
  },
  "message": "Voice clone created successfully"
}

Documentation Index

Fetch the complete documentation index at: https://docs.buena.ai/llms.txt

Use this file to discover all available pages before exploring further.

Create a new AI voice clone from audio samples. Upload high-quality audio files to train a custom voice model that can be used for LinkedIn voice messages and other voice synthesis applications.
Requires the voice:create permission. Maximum 5 voice clones per user.

​
Request

​
Authorization
string
required
Bearer token with your API key that has voice:create permission (e.g., “Bearer bna-your-api-key”)
​
Content-Type
string
required
Must be multipart/form-data for file uploads

​
Body Parameters

​
name
string
required
Descriptive name for the voice clone (e.g., “Professional Voice”, “Sales Persona”)
​
description
string
Optional description of the voice clone and its intended use
​
files
file[]
required
Audio files for voice training. Upload 1-10 high-quality audio files (WAV, MP3, M4A, FLAC). Each file must be under 50MB.
​
settings
object
Voice generation settings

​
Examples

​
Basic Voice Clone Creation

curl -X POST "https://api.buena.ai/api/v2/voice-clones" \
  -H "Authorization: Bearer bna-your-api-key" \
  -F "name=Professional Voice" \
  -F "description=Professional outreach voice for LinkedIn" \
  -F "files=@sample1.wav" \
  -F "files=@sample2.wav" \
  -F "files=@sample3.wav"

​
Advanced Voice Clone with Custom Settings

curl -X POST "https://api.buena.ai/api/v2/voice-clones" \
  -H "Authorization: Bearer bna-your-api-key" \
  -F "name=Sales Voice" \
  -F "description=Energetic sales voice for cold outreach" \
  -F "files=@recording1.wav" \
  -F "files=@recording2.wav"

​
Response

​
success
boolean
Always true for successful requests
​
data
object
The created voice clone information
{
  "success": true,
  "data": {
    "voiceId": "pNInz6obpgDQGcFmaJgB",
    "name": "Professional Voice",
    "description": "Professional outreach voice for LinkedIn",
    "sampleCount": 3,
    "createdAt": "2024-01-20T15:30:00.000Z"
  },
  "message": "Voice clone created successfully"
}

​
Voice Training Requirements

​
Audio Quality Guidelines

  • Supported formats: WAV, MP3, M4A, FLAC
  • Sample rate: 22,050 Hz or higher recommended
  • Bit depth: 16-bit minimum, 24-bit preferred
  • File size: Maximum 50MB per file
  • Duration: 30 seconds to 5 minutes per file optimal
  • Environment: Record in a quiet room with minimal echo - Microphone: Use a quality microphone, avoid phone recordings - Speaking style: Natural, conversational tone - Content variety: Include different emotions and sentence structures - Consistency: Same speaker, similar recording conditions
  • Language: Clear pronunciation and articulation
  • Variety: Mix of short and long sentences
  • Emotions: Include various emotional tones if desired
  • Avoid: Background noise, music, multiple speakers
  • Minimum: At least 1 minute of total audio recommended

​
Voice Clone Limits

Limit TypeValueNotes
Max voice clones per user5Contact support for higher limits
Max files per upload10Multiple uploads allowed
Max file size50MBPer individual file
Training time2-5 minutesDepends on audio length and quality
Supported languages29+Supported languages by Buena AI

​
Error Responses

​
Invalid File Format (400)

{
  "error": true,
  "code": "INVALID_FILE_FORMAT",
  "message": "Unsupported audio format. Please use WAV, MP3, M4A, or FLAC files.",
  "version": "2.0",
  "timestamp": "2024-01-20T15:30:00Z",
  "details": {
    "supported_formats": ["wav", "mp3", "m4a", "flac"],
    "rejected_files": ["recording.txt"]
  }
}

​
Voice Clone Limit Exceeded (429)

{
  "error": true,
  "code": "VOICE_CLONE_LIMIT_EXCEEDED",
  "message": "Maximum voice clones reached (5). Delete existing clones or upgrade your plan.",
  "version": "2.0",
  "timestamp": "2024-01-20T15:30:00Z",
  "details": {
    "current_count": 5,
    "max_allowed": 5,
    "upgrade_url": "https://app.buena.ai/billing"
  }
}

​
File Size Too Large (413)

{
  "error": true,
  "code": "FILE_TOO_LARGE",
  "message": "File size exceeds maximum limit of 50MB",
  "version": "2.0",
  "timestamp": "2024-01-20T15:30:00Z",
  "details": {
    "max_size_mb": 50,
    "rejected_files": [
      {
        "filename": "large_recording.wav",
        "size_mb": 75
      }
    ]
  }
}

​
Training Status Updates

Voice clone training is asynchronous. Monitor progress using: 
// Poll for status updates
async function checkTrainingStatus(voiceId) {
  const response = await fetch(`https://api.buena.ai/api/v2/voice-clones`, {
    headers: { Authorization: "Bearer bna-your-api-key" },
  });

  const { data } = await response.json();

  switch (data.status) {
    case "processing":
      console.log(`Training progress: ${data.training_progress}%`);
      break;
    case "ready":
      console.log("Voice clone ready for use!");
      break;
    case "failed":
      console.log("Training failed:", data.error_message);
      break;
  }

  return data;
}

​
Next Steps

List Voice Clones

View all your voice clones

Generate Preview

Test your voice clone with preview audio

LinkedIn Voice Messages

Send voice messages via LinkedIn

Voice Clone Management

Learn voice cloning best practices