The analysis and design of architecture systems for speech recognition on modern handheld-computing devices

Abstract

Growing demand for high performance in embedded systems is creating new opportunities to use speech recognition systems traditionally executed only on high performance systems. In several ways, the needs of embedded computing differ from those of more traditional general-purpose systems. Embedded systems have more stringent constraints on cost and power consumption that lead to design bottlenecks for many computationally-intensive applications. This paper characterizes the speech recognition process on hand-held mobile devices and evaluates the use of modern architecture features and compiler techniques for performing real-time speech recognition. We evaluate the University of Colorado Sonic speech recognition software on the IMPACT architectural simulator and compiler framework. Experimental results show that by using a strategic set of compiler optimization, a 500MHz processor with moderate levels of instruction-level parallelism and cache resources can meet the real-time computing and power constraints of an advanced speech recognition application.