# modelscope/ClearerVoice-Studio

An AI-Powered Speech Processing Toolkit and Open Source SOTA Pretrained Models, Supporting Speech Enhancement, Separation, and Target Speaker Extraction, etc.

Repository: https://github.com/modelscope/ClearerVoice-Studio
Canonical: https://ross.abutalabs.com/products/clearervoice-studio
Language: Python
License: Apache-2.0
License Family: permissive
Topics: audio, bandwidth-extension, deep-learning, noise-suppression, pytorch, speaker-extraction, speech, speech-enhancement, speech-separation, speech-super-resolution, speech-quality-evaluation
Last push: 2025-08-14T08:26:31+00:00

## Health v2 (maintenance only)
Score: 38/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 36, release rhythm 35, longevity 47
- inputs: {"age_days": 659, "days_push": 384, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4445, forks 364 (observed 2026-08-28T04:08:50.389957+00:00)

## What it is
ClearerVoice-Studio is an open-source, AI-powered speech processing toolkit from ModelScope/Alibaba offering state-of-the-art pretrained models for speech enhancement, separation, super-resolution, and target speaker extraction. It includes training and inference scripts, a pip-installable ClearVoice package, and a SpeechScore evaluation module.

## Use cases
- remove background noise from speech recordings
- separate overlapping speakers in an audio file
- extract a target speaker's voice from a mixture
- upsample low-quality audio to higher bandwidth speech
- evaluate speech quality with MOS-style metrics
- fine-tune speech enhancement models on custom data

## When to choose
- you need SOTA pretrained models for speech enhancement or separation
- you want both inference and training/fine-tuning scripts in one toolkit
- you need programmatic Python/NumPy integration into an audio pipeline

## When to avoid
- you need text-to-speech or speech recognition rather than enhancement
- you want a GUI desktop app rather than a Python library
- you lack GPU resources for fast inference on large models

## Facets
- artifact type: library
- maturity: active
- function: audio-processing, machine-learning, speech-recognition, deep-learning
- domain: speech-processing, machine-learning, deep-learning
- platform: python, cross-platform
- tags: speech-enhancement, speech-separation, noise-suppression, speech-super-resolution, speaker-extraction, pretrained-models, pytorch, speech-quality-evaluation, audio, gpu

## Member repositories
- modelscope/ClearerVoice-Studio (main) score 38

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:50.389957+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:20:48.638059+00:00, confidence not recorded.
  - readme: https://github.com/modelscope/ClearerVoice-Studio (fetched 2026-08-28T04:08:50.389957+00:00, sha 94ed50995aeb)
- Data as of 2026-08-30T08:39:29.467469+00:00.