Mobile CPU Based Optimization of Fast Likelihood Computation for Continuous Speech Recognition

We studied the real-time implementation of continuous speech recognition algorithm on a mobile CPU, Intel PXA270, platform. Especially, the optimization of fast likelihood computation, which takes the largest part in most speech recognizers, is conducted by employing SIMD (single instruction multiple data) programming and software pipelining. The overhead of exhaustive memory accesses is also minimized by placing frequently used acoustic model data at the fast internal SRAM. The number of execution cycles for the fast likelihood computation of the 1000-word vocabulary resource management (RM) task has been reduced by 56.42%. The resulting performance shows approximately four times faster processing speed than the real-time implementation requirement on a 520 MHz Intel XScale-based system.

[1]  Nikil D. Dutt,et al.  On-chip vs. off-chip memory: the data partitioning problem in embedded processor-based systems , 2000, TODE.

[2]  Hermann Ney,et al.  Using SIMD instructions for fast likelihood calculation in LVCSR , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[3]  J.H.L. Hansen,et al.  Fast likelihood computation techniques in nearest-neighbor based search for continuous speech recognition , 2001, IEEE Signal Processing Letters.

[4]  Ryosuke Isotani,et al.  Parallel LVCSR Algorithm for Cellphone-Oriented Multicore Processors , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[5]  Alexander I. Rudnicky,et al.  Pocketsphinx: A Free, Real-Time Continuous Speech Recognition System for Hand-Held Devices , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.