INTERACTIVE SOURCE MODELING FOR SPEECH


Koc T. , ÇİLOĞLU T.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 Nisan 2013 identifier identifier

Özet

In this study an interactive source model is proposed for voiced speech. The proposed model is compared with non-interactive model on glottal flow estimates obtained using closed-phase co-variance analysis. The results show that the interactive model is better than its non-interactive counterpart.