| Paper: | SLP-P13.8 | ||
| Session: | Speech Synthesis III | ||
| Time: | Thursday, May 18, 10:00 - 12:00 | ||
| Presentation: | Poster | ||
| Topic: | Speech and Spoken Language Processing: Signal Processing/Statistical Model for synthesis | ||
| Title: | HIGH QUALITY SINUSOIDAL MODELING OF WIDEBAND SPEECH FOR THE PURPOSES OF SPEECH SYNTHESIS AND MODIFICATION | ||
| Authors: | Dan Chazan, Ron Hoory, Ariel Sagi, Slava Shechtman, Alex Sorin, IBM Research Laboratory in Haifa, Israel, Israel; Zhi Wei Shuang, IBM China Research Lab, Israel; Raimo Bakis, IBM T. J. Watson Research Center, United States | ||
| Abstract: | This paper describes an efficient sinusoidal modeling framework for high quality wide band (WB) speech synthesis and modification. This technique may serve as a basis for speech compression in the context of small footprint concatenative Text to Speech systems. In addition, it is a useful representation for voice transformation and morphing purposes, e.g., simultaneous pitch modification and spectral envelope warping. The conventional sinusoidal modeling is enhanced with an adaptive frequency dithering mechanism, based on a degree of voicing analysis. Considerable reduction of the amount of model parameters is achieved by high band phase extension. The proposed model is evaluated and compared to the alternative STRAIGHT framework. While simpler and considerably more efficient than STRAIGHT, it outperforms it in speech quality for both speech reconstruction and transformation. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012