haoheliu/voicefixer

Health v2 · maintenance only

Adoption not part of the score

1373 stars · 163 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

VoiceFixer is a Python library and CLI tool for general speech restoration, using a pretrained neural vocoder to restore degraded human speech. It handles noise, reverberation, low resolution (2kHz-44.1kHz), and clipping within a single model.

Use cases

• restore degraded speech recordings
• denoise speech audio
• remove reverberation from voice recordings
• upsample low-resolution speech to 44.1kHz
• fix clipped audio recordings
• enhance old or low-quality voice recordings
• batch process wav files for speech enhancement

When to choose

• you need to restore seriously degraded speech in one model
• you want a pip-installable tool with CLI and Python API
• you need speech super-resolution, denoising, dereverberation, or declipping
• you want pretrained models without training your own

When to avoid

• you need real-time or streaming enhancement
• you need music or general audio restoration rather than speech
• you need to train a custom restoration model from scratch
• you need non-wav input formats without conversion

Facets

library · maturity stable

audio-processing speech-recognition machine-learning cli speech-processing machine-learning deep-learning python cli cross-platform speech-restoration denoising dereverberation declipping speech-super-resolution neural-vocoder speech-enhancement audio

Member repositories

For agents

markdown · JSON · MCP: product_card(name="haoheliu/voicefixer")