Effectiveness of A Direct Speech Transform Method Using Inductive Learning from Laryngectomee Speech to Normal Speech

摘要

This paper proposes and evaluates a new direct speech transform method with waveforms from laryngectomee speech to normal speech. Almost all conventional speech recognition systems and other speech processing systems are not able to treat laryngectomee speech with satisfactory results. One of the major causes is difficulty preparing corpora. It is very hard to record a large amount of clear and intelligible utterance data because the acoustical quality depends strongly on the individual status of such people. We focus on acoustic characteristics of speech waveform by laryngectomee people and transform them directly into normal speech. The proposed method is able to deal with esophageal and alaryngeal speech in the same algorithm. The method is realized by learning transform rules that have acoustic correspondences between laryngectomee and normal speech. Results of several fundamental experiments indicate a promising performance for real transform.