简体中文 | English
HitPawVoice is the official index repository for all audio and voice-related AI algorithms maintained by HitPaw-Official. This repository does not contain algorithm source code directly — it serves as a navigation hub with curated links, descriptions, and quick-start guides pointing to each standalone algorithm repository.
All algorithm repositories are independently versioned, documented, and deployable.
| Algorithm | Description | Repo Link | Paper |
|---|---|---|---|
| BeatDetection | Rhythm and beat detection from audio signals | HitPaw-Official/BeatDetection | — |
| SingingVoiceConversion | High-quality any-to-any singing voice style transfer | HitPaw-Official/SingingVoiceConversion | — |
Each algorithm repository under HitPaw-Official follows this layout:
AlgorithmName/
├── configs/ # Model and training configs
├── data/ # Dataset preparation scripts
├── deploy/ # Export, ONNX, and serving tools
├── docs/ # Extended documentation and assets
├── models/ # Model definitions
├── tools/ # Training / evaluation entry points
├── README.md
└── requirements.txt
Visit the specific algorithm repo for environment setup. All repos share a common pattern:
# 1. Clone the target algorithm repo
git clone https://github.com/HitPaw-Official/<AlgorithmName>.git
cd <AlgorithmName>
# 2. Install dependencies
pip install -r requirements.txt
# 3. Download pretrained weights (see each repo's README)
# 4. Run inference — see the specific algorithm repo's README for the exact command| Hub | Domain | Link |
|---|---|---|
| HitPawImage | Image AI algorithms | HitPaw-Official/HitPawImage |
| HitPawVideo | Video AI algorithms | HitPaw-Official/HitPawVideo |
We welcome contributions to any algorithm under HitPaw-Official. Please open issues or pull requests in the specific algorithm repository rather than this hub.
For organization-wide discussions, open an issue here.
All repositories under HitPaw-Official are released under the Apache 2.0 License unless otherwise noted in the individual repository.