Advances in objective voice quality assessment

The authors review the status of objective voice quality technology, including four newer techniques being considered by CCITT. Also included are results of applying the four methods to a speech dataset that represents three different languages and a variety of impairments. Data are used to compare and contrast the relative advantages and disadvantages of the assessment techniques. A discussion of problems involving with interpreting results of objective performance tests is presented. Finally, the relative reliability and accuracy of objective and subjective techniques are discussed and compared.<<ETX>>