A Smart Error Protection Scheme Based on Estimation of Perceived Speech Quality for Portable Digital Speech Streaming Systems

In this paper, a smart error protection (SEP) scheme is proposed to improve speech quality of a portable digital speech streaming (PDSS) system via a lossy transmission channel. To this end, the proposed SEP scheme estimates the perceived speech quality (PSQ) for received speech data, and then transmits redundant speech data (RSD) in order to assist speech decoder to reconstruct lost speech signals for high packet loss rates. According to the estimated PSQ, the proposed SEP scheme controls the RSD transmission, and then optimizes a bitrate of speech coding to encode the current speech data (CSD) against the amount of RSD without increasing transmission bandwidth. The effectiveness of the proposed SEP scheme is finally demonstrated using adaptive multirate-narrowband (AMR-NB) and ITU-T Recommendation P.563 as a scalable speech codec and a PSQ estimator, respectively. It is shown from experiments that a PDSS system employing the proposed SEP scheme significantly improves speech quality under packet loss conditions.