Automatic speech recognition system gain regulation involves computing upper, lower, mean energy traces for tracking speech signal, and computing normalized energy from upper trace

摘要

The method involves computing an upper energy trace for tracing the upper energy envelope of the speech signal; computing a lower energy trace for tracing the lower energy envelope of the speech signal; computing a middle energy trace for tracing the mean energy of the speech signal; and computing a normalized energy value from the upper energy trace for transfer to the speech recognition system. An Independent claim is also included for a speech recognition pre-processor.