Fuzzy expert system for continuous speech recognition

摘要

In designing a speech recognition expert system, there are a number of aspects to be considered—segmentation method, recognition unit, structure of the rule, and user aids. To avoid the difficulties in segmentation, an irregular unit based on spectral transition measure is proposed. Frames are used as the structure of the speech recognition rules. The structure provides the user with an easy way of making rules, and enables the building of an automatic rule generator. Fuzzy linguistic variables are used in representing rules. The rule-generating cycle, which includes a rule generator, a rule tester with an error reporter, and a state describer with other modules, could save much time in making rules and could result in high performance. The experimental results and comparison with the SPREX system, which is the basis of this system, are described.