Spectral conversion using deep neural networks trained with multi-source speakers

Abstract

This paper presents a method for voice conversion using deep neural networks (DNNs) trained with multiple source speakers. The proposed DNNs can be used in two ways for different scenarios: 1) in the absence of training data for source speaker, the DNNs can be treated as source-speaker-independent models and perform conversions directly from arbitrary… (More)
DOI: 10.1109/ICASSP.2015.7178892

Topics

4 Figures and Tables

Slides referencing similar topics