Opportunities and Challenges of Parallelizing Speech Recognition

TitleOpportunities and Challenges of Parallelizing Speech Recognition
Publication TypeConference Paper
Year of Publication2010
AuthorsChong, J., Friedland G., Janin A., Morgan N., & Oei C.
Conference NameSecond USENIX Workshop on Hot Topics in Parallelism (HotPar 2010)
Date Published6/2010

Automatic speech recognition enables a wide range of current and
emerging applications such as automatic transcription, multimedia
content analysis, and natural human-computer interfaces. This article
provides a glimpse of the opportunities and challenges that parallelism
provides for automatic speech recognition and related application
research from the point of view of speech researchers. The increasing
parallelism in computing platforms opens three major possibilities for
speech recognition systems: increasing recognition accuracy in
non-ideal, everyday noisy environments; increasing recognition
throughput in batch processing of speech data; and reducing recognition
latency in real-time usage scenarios. We explain the technical
challenges, the current approaches taken, and the possible directions
for future research on these three areas to guide the design of efficient
parallel software and hardware infrastructures.

icsi.pdf209.75 KB