Asynchronous Voice Cloning
The asynchronous voice cloning API is designed for processing longer text content that requires more time to generate. This approach is ideal for:
- Long-form content: Articles, books, or extensive documentation
- Large-scale processing: Multiple audio files or batch operations
- Background processing: Tasks that can run while users continue other activities
How It Works
The asynchronous API follows a three-step process:
- Create Task: Submit your audio sample and text for processing
- Monitor Progress: Check the status of your task periodically
- Download Result: Retrieve the generated audio when complete
API Endpoints
| Endpoint | Purpose | Description |
|---|---|---|
| Create Task | Submit processing request | Upload audio and text to start voice cloning |
| Task Status | Monitor progress | Check current status and estimated completion |
| Task Result | Download audio | Retrieve the generated voice clone |
When to Use Async API
Recommended for:
- Text longer than 1000 characters
- High-quality audio generation
- Batch processing multiple texts
- Applications that can handle delayed results
Processing Times:
- Short texts (1000-5000 chars): 2-5 minutes
- Medium texts (5000-15000 chars): 5-10 minutes
- Long texts (15000+ chars): 10-20 minutes
Getting Started
To begin using the asynchronous API:
- Prepare your content: Ensure your text is over 1000 characters
- Upload audio sample: Provide a 5-30 second voice sample
- Submit task: Use the Create Task endpoint
- Monitor progress: Poll the Task Status endpoint
- Download result: Retrieve your audio via Task Result
For shorter texts, consider using the Synchronous API for immediate results.
Last updated on