Speech Quality Evaluation: A New Application of Digital Watermarking

Speech quality evaluation is a very important research topic. In addition, the real time property of packet-switched network (e.g., the Internet) applications requires a fast and effective audio quality evaluation method. Mean opinion score (MOS) is reliable, but the listening test is very expensive, time-consuming, and sometimes impractical. The existing objective quality assessment methods require either original speech or a complicated computation model, which makes some applications of quality evaluation impossible. We propose to use digital audio watermarking to evaluate the quality of speech. The experimental results show that the method yields accurate quality scores that are very close to the results of PESQ