A training-based speech regeneration approach with cascading mapping models

Abstract

Computational speech reconstruction algorithms have the ultimate aim of returning natural sounding speech to aphonic and dysphonic patients as well as those who can only whisper. In particular, individuals who have lost glottis function due to disease or surgery, retain the power of vocal tract modulation to some degree but they are unable to speak anything more than hoarse whispers without prosthetic aid. While whispering can be seen as a natural and secondary aspect of speech communications for most people, it becomes the primary mechanism of communications for those who have impaired voice production mechanisms, such as laryngectomees.In this paper, by considering the current limitations of speech reconstruction methods, a novel algorithm for converting whispers to normal speech is proposed and the efficiency of the algorithm is explored. The algorithm relies upon cascading mapping models and makes use of artificially generated whispers (called whisperised speech) to regenerate natural phonated speech from whispers. Using a training-based approach, the mapping models exploit whisperised speech to overcome frame to frame time alignment problems that are inherent in the speech reconstruction process. This algorithm effectively regenerates missing information in the conventional frameworks of phonated speech reconstruction, and is able to outperform the current state-of-the-art regeneration methods using both subjective and objective criteria.