Abstract
An excellent reference and complete overview of distant automatic speech recognition: In the very recent past, automatic speech recognition (ASR) systems have attained acceptable performance when used with speech captured with a head-mounted or close-talking microphone (CTM). The performance of conventional ASR systems, however, degrades dramatically as soon as the microphone is moved away from the mouth of the speaker. This degradation is due to a broad variety of effects that are not found in CTM speech, including background noise, overlapping speech from other speakers, and reverberation. While conventional ASR systems underperform for speech captured with far-field sensors, there are a number of techniques developed in other areas of signal processing that can mitigate the deleterious effects of noise and reverberation, as well as separating speech from overlapping speakers. Distant Speech Recognition presents a contemporary and comprehensive description of both theoretic abstraction and practical issues inherent in the distant ASR problem.
