Machine Hearing: An Emerging Field

If we had machines that could hear as humans do, we would expect them to be able to easily distinguish speech from music and background noises, to pull out the speech and music parts for special treatment, to know what direction sounds are coming from, to learn which noises are typical and which are noteworthy.

Hearing machines should be able to listen and react in real time, to take appropriate action on hearing noteworthy events, to participate in ongoing activities.

In the September issue of IEEE Signal Processing Magazine, Richard F. Lyon’s “Exploratory DSP” column article introduces the filed of machine hearing. Applications of machine hearing are abundant and many are easy to address with known auditory front ends, combined with known feature extraction and machine learning techniques such as those that have proven successful in analogous applications in machine vision. There is still room for open-ended research and improvement. It is stated that “cooperation with researchers in auditory psychology and physiology will be highly valued on both ends.”