Speaking-Aid Systems Based on One-to-Many Eigenvoice Conversion for Total Laryngectomees

This paper proposes speaking-aid systems based on one-to-many eigenvoice conversion (EVC) for enhancing three types of alaryngeal speech: esophageal speech; electrolaryngeal speech; and body-conducted silent electrolaryngeal speech. Although alaryngeal speech allows laryngectomees to utter speech sounds, it suffers from lack of naturalness and speaker individuality. To improve the sound quality of alaryngeal speech, alaryngeal-speech-to-speech (AL-to-Speech) methods based on statistical voice conversion have been proposed. To recover the speaker individuality of alaryngeal speech, a one-to-many EVC capable of flexibly adapting the conversion model to given target natural voices was applied to the AL-to-Speech methods. The experimental results of objective and subjective evaluations demonstrate that the proposed methods yield significant improvements of speech quality and make the converted voice quality similar to the given target voice quality.