Generative Moment Matching Network-Based Neural Double-Tracking for Synthesized and Natural Singing Voices

Hiroki TAMARU  Yuki SAITO  Shinnosuke TAKAMICHI  Tomoki KORIYAMA  Hiroshi SARUWATARI  

IEICE TRANSACTIONS on Information and Systems   Vol.E103-D   No.3   pp.639-647
Publication Date: 2020/03/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2019EDP7228
Type of Manuscript: PAPER
Category: Speech and Hearing
DNN-based singing-voice synthesis,  generative moment matching network,  inter-utterance pitch variation,  artificial double-tracking,  modulation spectrum,  

Full Text: PDF(893.2KB)>>
Buy this Article

This paper proposes a generative moment matching network (GMMN)-based post-filtering method for providing inter-utterance pitch variation to singing voices and discusses its application to our developed mixing method called neural double-tracking (NDT). When a human singer sings and records the same song twice, there is a difference between the two recordings. The difference, which is called inter-utterance variation, enriches the performer's musical expression and the audience's experience. For example, it makes every concert special because it never recurs in exactly the same manner. Inter-utterance variation enables a mixing method called double-tracking (DT). With DT, the same phrase is recorded twice, then the two recordings are mixed to give richness to singing voices. However, in synthesized singing voices, which are commonly used to create music, there is no inter-utterance variation because the synthesis process is deterministic. There is also no inter-utterance variation when only one voice is recorded. Although there is a signal processing-based method called artificial DT (ADT) to layer singing voices, the signal processing results in unnatural sound artifacts. To solve these problems, we propose a post-filtering method for randomly modulating synthesized or natural singing voices as if the singer sang again. The post-filter built with our method models the inter-utterance pitch variation of human singing voices using a conditional GMMN. Evaluation results indicate that 1) the proposed method provides perceptible and natural inter-utterance variation to synthesized singing voices and that 2) our NDT exhibits higher double-trackedness than ADT when applied to both synthesized and natural singing voices.