haoheliu/voicefixer

General Speech Restoration

64
/ 100
Established

Combines a neural vocoder-based restoration module with a speaker-independent synthesis vocoder to handle multiple degradation types—noise, reverberation, clipping artifacts, and bandwidth limitations (2kHz–44.1kHz)—within a single unified model. Supports three operating modes (original, preprocessing-enhanced, and training-focused) and provides CLI, Python API, and Streamlit web interfaces with optional GPU acceleration via CUDA.

1,302 stars and 2,551 monthly downloads. No commits in the last 6 months. Available on PyPI.

Stale 6m
Maintenance 0 / 25
Adoption 18 / 25
Maturity 25 / 25
Community 21 / 25

How are scores calculated?

Stars

1,302

Forks

156

Language

Python

License

MIT

Last pushed

Feb 17, 2025

Monthly downloads

2,551

Commits (30d)

0

Dependencies

8

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/voice-ai/haoheliu/voicefixer"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.