Esophageal speech enhancement using source synthesis and formant patterns modification

Summary

This paper deals with esophageal speech, which is a voice of substitution used by alaryngeal persons in order to be able to communicate with others. This voice, characterized by a low intensity and poor intelligibility, is hard to understand.In this paper, we propose ideas to enhance this kind of voice. More precisely, we enhance the source excitation signal and the formant structure of the speech vocal track. We modify pitch values by those of natural speech and we replace the source by a synthetic one based on LF model. We also enhance the formant structure by enlarging formant bandwidth and we amplify their amplitudes without increasing the background noise. Then, we englobe all modification in the same scheme including all improvements.