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

Li-Juan Liu; Ling-Hui Chen; Zhen-Hua Ling; Li-Rong Dai

ICASSP 2015accepted0 citations

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

Li-Juan Liu, Ling-Hui Chen, Zhen-Hua Ling, Li-Rong Dai

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 source speakers to certain target speaker; 2) the DNNs can also be used as initial models for further fine-tuning of source-speaker-dependent DNNs when parallel training data for both source and target speakers are available. Experimental results show that, as source-speaker-independent models, the proposed DNNs can achieve comparable performance to conventional source-speaker-dependent models. On the other hand, the proposed method outperforms the conventional initialization method with restricted Boltzmann machines (RBMs).

BibTeX

@inproceedings{icassp2015_spectralconversi,
  title = {Spectral conversion using deep neural networks trained with multi-source speakers},
  author = {Li-Juan Liu and Ling-Hui Chen and Zhen-Hua Ling and Li-Rong Dai},
  booktitle = {ICASSP 2015},
  year = {2015}
}