A Method of Subject Extension Pitch Extraction for Humming and Singing Signals

DOI

10.2991/cset-16.2016.71How to use a DOI?

Keywords

Humming query, pitch extraction, endpoint detection, subject extension

Abstract

Pitch extraction is a key task of ahumming query system.The purpose of this paper isto find a method to extract pitch accuratelyfortwo input modes, humming and singing.According to the characteristics of Chinese pronunciation, this paper presentsa new method, namely the Pitch Extraction of Subject Extension.According to the differencesbetween humming and singing signals, this method respectively chooses energy threshold and ratio of energy and entropy to detect endpoint of notes.The candidate method and the shortest distance method are used to determine the pitch periods of the voiced segments, and subject extension method is used to determine the pitch periods of the voiced/unvoiced mixed segments. Finally this algorithm is implemented in a small database, andis compared withother similar algorithms. Experiments show our algorithm is more accurate and robust.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Jinghui Zhang
AU  - Shen Yang
AU  - Huahua Wu
PY  - 2016/08
DA  - 2016/08
TI  - A Method of Subject Extension Pitch Extraction for Humming and Singing Signals
BT  - Proceedings of the 2016 International Conference on Computer Science and Electronic Technology
PB  - Atlantis Press
SP  - 304
EP  - 309
SN  - 2352-538X
UR  - https://doi.org/10.2991/cset-16.2016.71
DO  - 10.2991/cset-16.2016.71
ID  - Zhang2016/08
ER  -