Abstract

In this paper, a robust distant talking speech recognition system for a home automation system application is presented based on a microphone array. We use commands such as turning on/off or openinglclosing for controlling diferent device in a home or office. By combining diferent enhancement methods such as sub-array beamforming, spectrum enhancement, feature domain enhancement we increase the word recognition rate of isolated speech recognition system in adverse noisy condition. Two databases is recorded for this reason first database is recorded with a close-talk microphone for training the system and for testing the system second noisy database is recorded in real environment. We also investigate the effect of adding each enhancement method on accuracy of the system. In another experiment we use part of the noisy data for training and another part for testing the system and reporting the results.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call