Lyric recognition in monophonic singing using pitch-dependent DNN

Dairoku Kawai; Kazumasa Yamamoto; Seiichi Nakagawa

ICASSP 2017accepted0 citations

Lyric recognition in monophonic singing using pitch-dependent DNN

Dairoku Kawai, Kazumasa Yamamoto, Seiichi Nakagawa

Abstract

One of the difficulties in sung speech recognition is the small distance in an acoustic space between phonemes in sung speech. Therefore we considered clustering the speech based on a pitch (fundamental frequency F0) and creating a larger distance between the phonemes. In addition, we considered a two-stage training method of DNN-HMM: the first stage is trained by using conventional acoustic features like MFCCs, and the second stage is re-trained by augmented features with a pitch feature. We expected to train pitch information more explicitly in the second stage of the training and obtained a relative improvement of 9% as expected.

BibTeX

@inproceedings{icassp2017_lyricrecognition,
  title = {Lyric recognition in monophonic singing using pitch-dependent DNN},
  author = {Dairoku Kawai and Kazumasa Yamamoto and Seiichi Nakagawa},
  booktitle = {ICASSP 2017},
  year = {2017}
}