HVC-Audio-Convert Base Models

Overview

These models serve as the foundational components for HVC-Audio-Convert (Soft-VC Voice Conversion), an advanced voice conversion framework that combines SoftVC feature extraction with the VITS (Conditional Variational Autoencoder with Adversarial Learning) architecture.

Key Features

  • High-quality voice conversion capabilities
  • Pre-trained on diverse vocal datasets
  • Supports cross-lingual voice conversion
  • Compatible with HVC-Audio-Convert v4.0 and newer

Technical Details

  • Architecture: Based on VITS (Conditional Variational Autoencoder)
  • Feature Extraction: Hibernates content encoder
  • Training Data: Curated multi-speaker datasets
  • Model Format: PyTorch checkpoints

Usage

  1. Download the desired base model
  2. Use with HVC-Audio-Convert framework
  3. Fine-tune on target voice data
  4. Perform voice conversion

Requirements

  • HVC-Audio-Convert framework
  • Python 3.8+
  • PyTorch 1.13.0+
  • CUDA compatible GPU (recommended)

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Citation

If you use these models in your research, please cite:

Downloads last month
0
Inference Providers NEW
This model is not currently available via any of the supported third-party Inference Providers, and HF Inference API was unable to determine this model's library.