https://github.com/resemble-ai/resemble-enhance

AI powered speech denoising and enhancement
https://github.com/resemble-ai/resemble-enhance

denoise speech-denoising speech-enhancement speech-processing

Last synced: 24 days ago
JSON representation

AI powered speech denoising and enhancement

• Host: GitHub
• URL: https://github.com/resemble-ai/resemble-enhance
• Owner: resemble-ai
• License: mit
• Created: 2023-11-15T08:15:51.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2024-06-21T01:28:13.000Z (5 months ago)
• Last Synced: 2024-10-01T00:41:02.858Z (about 1 month ago)
• Topics: denoise, speech-denoising, speech-enhancement, speech-processing
• Language: Python
• Homepage: https://huggingface.co/spaces/ResembleAI/resemble-enhance
• Size: 20.5 KB
• Stars: 1,311
• Watchers: 18
• Forks: 133
• Open Issues: 40
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• StarryDivineSky - resemble-ai/resemble-enhance

README

        
# Resemble Enhance

[![PyPI](https://img.shields.io/pypi/v/resemble-enhance.svg)](https://pypi.org/project/resemble-enhance/)

[![Hugging Face Space](https://img.shields.io/badge/Hugging%20Face%20%F0%9F%A4%97-Space-yellow)](https://huggingface.co/spaces/ResembleAI/resemble-enhance)

[![License](https://img.shields.io/github/license/resemble-ai/Resemble-Enhance.svg)](https://github.com/resemble-ai/resemble-enhance/blob/main/LICENSE)

[![Webpage](https://img.shields.io/badge/Webpage-Online-brightgreen)](https://www.resemble.ai/enhance/)

https://github.com/resemble-ai/resemble-enhance/assets/660224/bc3ec943-e795-4646-b119-cce327c810f1

Resemble Enhance is an AI-powered tool that aims to improve the overall quality of speech by performing denoising and enhancement. It consists of two modules: a denoiser, which separates speech from a noisy audio, and an enhancer, which further boosts the perceptual audio quality by restoring audio distortions and extending the audio bandwidth. The two models are trained on high-quality 44.1kHz speech data that guarantees the enhancement of your speech with high quality.

## Usage

### Installation

Install the stable version:

```bash

pip install resemble-enhance --upgrade

```

Or try the latest pre-release version:

```bash

pip install resemble-enhance --upgrade --pre

```

### Enhance

```

resemble_enhance in_dir out_dir

```

### Denoise only

```

resemble_enhance in_dir out_dir --denoise_only

```

### Web Demo

We provide a web demo built with Gradio, you can try it out [here](https://huggingface.co/spaces/ResembleAI/resemble-enhance), or also run it locally:

```

python app.py

```

## Train your own model

### Data Preparation

You need to prepare a foreground speech dataset and a background non-speech dataset. In addition, you need to prepare a RIR dataset ([examples](https://github.com/RoyJames/room-impulse-responses)).

```bash

data

├── fg

│   ├── 00001.wav

│   └── ...

├── bg

│   ├── 00001.wav

│   └── ...

└── rir

    ├── 00001.npy

    └── ...

```

### Training

#### Denoiser Warmup

Though the denoiser is trained jointly with the enhancer, it is recommended for a warmup training first.

```bash

python -m resemble_enhance.denoiser.train --yaml config/denoiser.yaml runs/denoiser

```

#### Enhancer

Then, you can train the enhancer in two stages. The first stage is to train the autoencoder and vocoder. And the second stage is to train the latent conditional flow matching (CFM) model.

##### Stage 1

```bash

python -m resemble_enhance.enhancer.train --yaml config/enhancer_stage1.yaml runs/enhancer_stage1

```

##### Stage 2

```bash

python -m resemble_enhance.enhancer.train --yaml config/enhancer_stage2.yaml runs/enhancer_stage2

```

## Blog

Learn more on our [website](https://www.resemble.ai/enhance/)!