New approaches for domain transformation and parameter combination for improved accuracy in parallel model combination (PMC) techniques

Parallel model combination (PMC) techniques have been very successful and p opularly used in many applications to improve the performance of speech rec ognition systems under noisy environments. However, it is believed that som e assumptions and approximations made in this approach, primarily in the do main transformation and parameter combination processes, are not necessaril y accurate enough in certain practical situations, which may degrade the ac hievable performance of PMC. In this paper, the possible sources that cause the performance degradation in these processes are carefully analyzed and discussed. Three new approaches, including the truncated Gaussian approach and the split mixture approach for domain transformation process and the es timated cross-term approach for parameter combination process, are proposed in this paper in order to handle these problems, minimize such degradation , and improve the accuracy of the PMC techniques. These proposed approaches were analyzed and discussed with two recognition tasks, one relatively sim ple, and the other more complicated and realistic. Both sets of experiments showed that these proposed approaches are able to provide significant impr ovements over the original PMC method, especially when the SNR condition is worse.