Testing the limits: Quantifying the degradation of automatic speech recognition in reverberant environments.

Abstract

Surprisingly little is known about the specific character of the depreciation of automatic speech recognition in reverberation—its primary acoustics causes (e.g., room geometry and reverberation strength) or speech effects (blurring of syllables, plosives, and consonants). The focus of this study is to precisely quantify the depreciation of speech recognition accuracy for reverberant signals using a black box experiment to vary reverberation characteristics and observe speech recognition accuracy. The methodology tests two speech recognition platforms on a recognition task of similar sounding word lists. A range of reverberant settings was simulated by convolution with an impulse response. The recognizers had the least reverberant recognition accuracy for words which only differed by their ending consonants. The depreciation of recognition accuracy from early reflections alone was lower than the overall room effect; however, the overall depreciation with respect to the absorption coefficient was well predicted by the strength of the reverberant tail. The results were compared to the results of prior research.