Pitch Prediction from Mel-generalized Cepstrum - a Computationally Efficient Pitch Modeling Approach for Speech Synthesis

Rao, Achuth M and Ghosh, Prasanta Kumar (2017) Pitch Prediction from Mel-generalized Cepstrum - a Computationally Efficient Pitch Modeling Approach for Speech Synthesis. In: 25th European Signal Processing Conference (EUSIPCO), AUG 28-SEP 02, 2017, GREECE, pp. 1629-1633.

PDF Eur_Sig_Pro_Con_1629_2017.pdf - Published Version Restricted to Registered users only Download (417kB) | Request a copy

Abstract

Text-to-speech (TTS) systems are often used as part of the user interface in wearable devices. Due to limited memory and computational/battery power in wearable devices, it could be useful to have a TTS system which requires less memory and is less computationally intensive. Conventional speech synthesis systems has separate modeling for pitch (F0-model) and spectral representation, namely Mel generalized coefficients (MGC) (MGC-model). In this paper we estimate pitch from the MGC estimated using MGC-model instead of having a separate F0-model. Pitch is obtained from the estimated MGC using a statistical mapping through Gaussian mixture model (GMM). Experiments using CMU-ARCTIC database demonstrate that the proposed GMM based F0-model, even with a single mixture, results in no significant loss in the naturalness of the synthesized speech while the proposed F0-model, in addition to reducing computational complexity, results in similar to 93% reduction in the number of parameters compared to that of the F0-model.

Item Type:	Conference Proceedings
Series.:	European Signal Processing Conference
Publisher:	IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA
Additional Information:	Copy right for this article belong to IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA
Department/Centre:	Division of Electrical Sciences > Electrical Engineering
Date Deposited:	18 Apr 2018 18:22
Last Modified:	18 Apr 2018 18:22
URI:	http://eprints.iisc.ac.in/id/eprint/59638

Actions (login required)

View Item