Title :
Objective and subjective optimization of APC system performance
Author :
Gibson, Jerry D. ; Chang, Wen-Whei
Author_Institution :
Dept. of Electr. Eng., Texas A&M Univ., College Station, TX, USA
fDate :
6/1/1990 12:00:00 AM
Abstract :
System configurations and relative performance results are presented for adaptive predictive coding (APC) systems at 9.6 kb/s based on three different criteria: subjective listening tests; signal-to-quantization noise ratio (SNR); and segmental SNR (SNRseg). The subjectively optimized coder has little granular noise but some spectral distortion and SNR and SNRseg values less than half those of the other two systems. The objectively optimized systems have significant granular noise but a better reproduction of the signal spectrum. Optimization of SNRseg does not lead to the subjectively preferred design
Keywords :
encoding; interference (signal); speech analysis and processing; speech intelligibility; 9.6 kbit/s; APC system performance; adaptive predictive coding; granular noise; objectively optimized systems; segmental SNR; signal spectrum reproduction; signal/quantisation noise ratio; spectral distortion; speech encoding; subjective listening tests; subjective optimization; Filters; Frequency response; Predictive coding; Quantization; Random variables; Roundoff errors; Signal to noise ratio; Speech processing; System performance; Upper bound;
Journal_Title :
Acoustics, Speech and Signal Processing, IEEE Transactions on
