H91-1031 | reduce run-time computation for | vowel classification | with no cost to accuracy . Finally |
H91-1031 | feature values are combined for | vowel classification | . In some conditions ( cf. conditions |
H91-1027 | of multilayer peroeptron based | vowel classification | . The dual auditory model ( mean |
H91-1031 | similar performance to direct | vowel classification | , but potentially offers a more |