How can I make realtime filter of specific sounds in my TV?
Hi forum! I noticed that my dog goes mad when it hears another dog barking in a TV show. I am curious, is it possible to make a real-time filter for my TV that will remove those sounds?
See below possible solutions: - Acquire a dataset of 'clean' audio samples without barking dogs and a dataset of barking dogs sounds. - Generate a training set by mixing random selection of clean audio with a random selection of barking dog noises at various volume levels. This is your input data. Your output data is the clean audio before you added the barking. - Find a suitable neural network. This might be a good starting point: https://github.com/andabi/music-source-separation - Train the network. - Apply the network to a load of the clean+bark mixes and check the result by ear. If the results are not good enough, go to 3 to tweak the network. If the results are still poor go to 2 and generate more training data, perhaps adding some distortions or other effects the barking.