RNNoise: Learning Noise Suppression

[jmvalin]

This demo presents the RNNoise project, showing how deep learning can be applied to noise suppression. The main idea is to combine classic signal processing with deep learning to create a real-time noise suppression algorithm that's small and fast. No expensive GPUs required — it runs easily on a Raspberry Pi. The result is much simpler (easier to tune) and sounds better than traditional noise suppression systems (been there!).

Read More

Comments

i suggest addition option to open local file

[(Anonymous)]

i suggest addition option to open local file

RNNoise running demo issue

[(Anonymous)]

Dear Jean-Marc,
I'm trying to run compiled rnnoise demo on windows 10 using command:
./examples/rnnoise_demo input.wav output.wav
Why am I getting an empty audio file at the output (output.wav 0 KB)?
Thanks in advance. Anton

Re: RNNoise running demo issue

[(Anonymous)]

Fixed, it was just a file directory conflict.

Re: RNNoise running demo issue

[(Anonymous)]

We can put any type of file as input? and where are you placing the inpit file? Inside the root directory or the examples directory.

[(Anonymous)]

Hey Jean-Marc,

I don't know much about all this neural network stuff, but I know this project is wicked cool and works well.

Could RNNoise be adapted to run on the GNA processor built in to Intel's latest CPUs?

The lib, API, and a sample source are at https://github.com/intel/gna if you're interested in taking a look.

Thanks so much for sharing this work, and a belated congrats for the Open Broadcaster Software integration :)

- B

Unable to clear the background noise using this repo even after using the correct PMC files

[(Anonymous)]

It is returning a big output file that's it