haoheliu/versatile_audio_super_resolution

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1951 stars · 201 forks observed · 2026-08-28

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

AudioSR is a Python library and CLI tool that performs versatile audio super-resolution, upsampling any audio (music, speech, sound effects) to 48kHz using diffusion-based models. It offers basic and speech-specific checkpoints, a Gradio demo, and batch processing of audio files.

Use cases

• upsample low-quality audio to 48khz
• enhance resolution of music recordings
• improve speech audio fidelity
• batch super-resolve a list of audio files
• restore old or compressed audio samples
• super-resolve sound effects for games or film

When to choose

• you need to increase the sampling rate and fidelity of arbitrary audio types
• you want a ready-to-use CLI or pip-installable model rather than building your own
• you have GPU resources and want state-of-the-art diffusion-based audio enhancement

When to avoid

• you need real-time or low-latency enhancement
• you lack a GPU and cannot tolerate slow inference
• you only need simple resampling without fidelity enhancement (use ffmpeg instead)

Facets

library · maturity active

audio-processing machine-learning cli machine-learning media python cli cross-platform audio-super-resolution diffusion-models speech-enhancement music-enhancement 48khz gradio-demo audio gpu

Member repositories

For agents

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