计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (16): 124-126.DOI: 10.3778/j.issn.1002-8331.2010.16.036

• 数据库、信号与信息处理 • 上一篇    下一篇

歌词到歌曲转换系统的实现

李锦珑,杨鸿武,梁青青,裴 东,刘慧娟   

  1. 西北师范大学 物理与电子工程学院,兰州 730070
  • 收稿日期:2009-11-10 修回日期:2010-02-04 出版日期:2010-06-01 发布日期:2010-06-01
  • 通讯作者: 李锦珑

Lyrics to singing voice synthesis system

LI Jin-long,YANG Hong-wu,LIANG Qing-qing,PEI Dong,LIU Hui-juan   

  1. College of Physics and Electronic Engineering,Northwest Normal University,Lanzhou 730070,China
  • Received:2009-11-10 Revised:2010-02-04 Online:2010-06-01 Published:2010-06-01
  • Contact: LI Jin-long

摘要: 将文语转换技术与语音修改技术相结合,实现了一个歌词到歌曲的转换系统。首先利用一个文语转换系统将输入的歌词转换为语音,同时从歌曲的MIDI文件中提取歌曲的旋律参数,最后通过旋律控制模型对语音信号的声学特征进行修改,实现由歌词到歌曲的转换。实验结果表明,系统合成的歌曲达到了3.29的平均MOS得分。

关键词: 文语转换, 歌曲合成, 旋律控制模型, 声学特征, 乐器数字接口(MIDI)

Abstract: The text to speech synthesis technology and the speech modification technique are combined together to convert the lyrics to song.Firstly,the inputs of the lyrics are converted into speech with the text to speech synthesis system.Then,the melody parameters of the song are extracted from the MIDI file.At last,the melody control model is adopted to achieve the lyrics to song conversion by modifying the acoustic parameters of speech signal.Experiments demonstrate that the system can achieve 3.29 of average MOS score.

Key words: text to speech synthesis, singing voice synthesis, melody control model, acoustic feature, Musical Instrument Digital Interface(MIDI)

中图分类号: