I have searched and found a lot of noise-canceling solutions. but many of them are platform-dependent(like "noise torch" for ubuntu) and many of them are not workable for real-time. I also looked into some RNN(recurrent neural network) based solution but none of this is that good or hard to integrate.
now, for an audio conference situation, especially targeted to an office environment(any formal conference mainly), what approach should I follow for denoising audio(audio byte array for example) per participant? or what is the best open-source (may need some tweaking) solution/approach available to date which is platform-independent?
Thanks in advance for any kind of help..!!