Embedded speech recognition system on 8-bit MCU core

Author

Wang, Dong ; Zhang, Liang ; Liu, Jia ; Liu, Runsheng

Author_Institution

Dept. of Electron. Eng., Tsinghua Univ., Beijing, China

Volume

5

fYear

2004

fDate

17-21 May 2004

Abstract

This is a small vocabulary, speaker independent, discrete word speech recognition system based on the system on chip (SOC) philosophy. It is implemented on an 8-bit MCU (micro control unit). The system adopts linear predictive cepstral coefficient (LPCC) related features followed by a vector quantization (VQ) step as the front end, and a hidden Markov model (HMM) as the speech model. Confidence measures based on likelihood scores (LLS) are given for rejection of out-of-vocabulary (OOV) words. The recognition rate is improved with corrective training, and robustness is acquired by integrating the confidence measure into the system. The recognition accuracy is nearly 97% with a vocabulary up to 30 phrases under normal conditions. A simple speech codec is also implemented for all speech I/O purposes.

Keywords

cepstral analysis; embedded systems; hidden Markov models; microcontrollers; speech codecs; speech recognition equipment; system-on-chip; vector quantisation; 8 bit; LPCC; MCU core; SOC; corrective training; discrete word speech recognition; embedded speech recognition system; hidden Markov speech model; likelihood score confidence measure; linear predictive cepstral coefficient related features; out-of-vocabulary word rejection; recognition accuracy; recognition rate; small vocabulary speaker independent speech recognition; speech I/O codec; vector quantization; Cepstral analysis; Hidden Markov models; Predictive models; Robustness; Semiconductor device measurement; Speech codecs; Speech recognition; System-on-a-chip; Vector quantization; Vocabulary;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2004. Proceedings. (ICASSP '04). IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-8484-9

Type

conf

DOI

10.1109/ICASSP.2004.1327107

Filename

1327107