Effects of channel conditions on the performance of ASR over GSM networks using full rate speech codec

Author

Reddy, M. Ram ; Mangamma, S. Alivelu ; Laxminarayana, P. ; Ramana, A.V. ; Gangadhar, P.

Author_Institution

NERTU, Osmania Univ., Hyderabad, India

fYear

2014

Firstpage

1

Lastpage

6

Abstract

In this paper, the performance of Automatic Speech Recognition (ASR) and Mean Opinion Score (MOS) are evaluated under different channel conditions over GSM Network using standard narrow band Full Rate speech and channel coding algorithms. CMU SPHINX toolkit and TIMIT speech database (8 kHz raw files) are used to evaluate the ASR accuracy. MOS values are estimated, using PESQ P.862, as recommended by ITU-T (International Telecommunication Union). In this evaluation, Rayleigh distribution for generation of channel noise is used to simulate the channel conditions. The results obtained in the experiments are reported and analyzed.

Keywords

Rayleigh channels; cellular radio; channel coding; speech codecs; speech recognition; ASR accuracy; CMU SPHINX toolkit; GSM networks; ITU-T; MOS values; PESQ P.862; Rayleigh distribution; TIMIT speech database; automatic speech recognition; channel coding algorithms; channel conditions; channel noise generation; full rate speech codec; international telecommunication union; mean opinion score; standard narrow band; Bit error rate; Channel coding; GSM; Noise; Speech; Speech coding; Speech recognition; ASR; CMU SPHINX; GMSK; GSM-FR; MOS; Rayleigh fading channel model; TIMIT data base;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer and Communications Technologies (ICCCT), 2014 International Conference on

Type

conf

DOI

10.1109/ICCCT2.2014.7066729

Filename

7066729