RNNoise is a noise suppression library based on a recurrent neural network

Quick Demo application

While it is meant to be used as a library, a simple command-line tool is provided as an example. Here is a Colab Demo to Convert Raw Noised Audio to Cleaned Audio.

build librnnoise & rnnoise_demo with CMake

# mkdir build
# cd build
# cmake ..
# make

It operates on wav and mp3 files, which can be used as:

# ./rnnoise_demo input.wav
# ./rnnoise_demo input.mp3

the output filename is "input_out.wav" or:

specify the output filename

# ./rnnoise_demo input.wav output.wav
# ./rnnoise_demo input.mp3 output.wav

References and Resources:

• david8862/rnnoise
• RNNoise: Learning Noise Suppression
• RNNoise: Learning Noise Suppression（深度学习噪声抑制）
• 基于RNN的音频降噪算法
• ZhiHan Gao

License

This work is licensed under MIT License. See LICENSE for details.

Donating

If you found this project useful, consider buying me a coffee