Enhancing Objective Evaluation of Speech Quality Algorithm: Current Efforts, Limitations and Future Directions

We discuss a performance improvement over the current state- of-the-art objective speech quality assessment algorithm by means of a parameter study. The algorithm, PESQ, computes disturbances surfaces, and applies a cognitive model, which is simply a regression. We note that many parameters appear to be tuned arbitrarily, and set them with a more systematic method. We search for optimum Lp-norms over the frequency-time domain disturbance surface. The optimum Lp-norms yield the most desirable correspondence between symmetric/asymmetric disturbance terms and subjective scores. New features are added to the cognitive processing algorithm. Performance is improved by a wide margin over the standard PESQ. However, the results are still not satisfying for the intended goals of the algorithm. Limitations and bottlenecks of the current standardized approaches, as well as emerging new ideas, are reviewed. Finally, we call for more innovative, rather than renovative, research efforts in objective speech quality measurement for fundamental enhancement of the algorithm.