acm-header
Sign In

Communications of the ACM

ACM TechNews

Speaking in Song


We want to use our technology to help the average person sing well," the researcher says.

Voice synthesis software developed at Singapores Agency for Science, Technology and Research will make you sound like the melodious singer you have always wanted to be.

Credit: Sergey Nivens

Researchers from Singapore's Agency for Science, Technology, and Research (A*STAR) have developed I2R Speech2Singing, voice synthesis software they say can make anyone's singing voice sound more melodious.

The software automatically produces high-quality singing while preserving the original character of the user's natural voice.

The program works by polishing melody while retaining the original content and timbre of a sound. It uses recordings by professional singers as templates to correct the melody of a singing voice or to convert a speaking voice into a singing one.

The software detects the timing of each phonetic sound using speech-recognition technology and then stretches or compresses the duration of the signal using voice-conversion technology to match the rhythm to that of a professional singer. A speech synthesizer then combines the time-corrected voice with pitch data and background music to produce a melodic solo.

"When we compared the output with other currently available applications, we realized that our software generated a much better voice quality," says A*STAR's Minghui Dong.

The researchers now are developing a way to quickly add songs to the software so large-scale song databases can be built.

From ResearchSEA
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found