Friday Decenber 5, 2003
UPGRADE TO NEW ATLAS PLUS
More than 1,500 New Atlas Plus subscribers directly support our journalism, and get access to our premium ad-free site and email newsletter. Join them for just US$19 a year.UPGRADE
The voice as musical instrument in the computer age
VOCALOID is an impressive new Singing Synthesis Software developed by YAMAHA that generates authentic singing sounds using lyrics and notes entered into a computer by the composer - meaning that full voice parts can be created without the need for a singer.
Since the earliest days of computerised speech artificially generated voices have always sounded, well, like artificially generated voices. Despite steady improvement over the past decade (more recent text to speech programs even offer a wide selection of different accents that the computer can use), the quality is voice synthesis achieved by VOCALOID has reached a new level that's superior to anything we've heard to date.
UK based digital audio developer Zero-G Limited is working with YAMAHA to develop the first high quality vocal libraries in English. The recordings of actual singing provide the palette from which the voice synthesis is created and composers can also add expressive effects such as vibrato and pitch bends using the Windows-based system.
To synthesize vocal parts, the system retrieves data consisting of voice snippets, applies pitch conversion, and splices and shapes them to form the lyrics of the song entered by the composer. During this process pitch can easily changed according to the melody and the voice snippets can be spliced to generate smooth-flowing words.
The result is a realistic synthesised voice that retains the vocal qualities of the original singing and has a far greater range of lyrical and vocal expression than can be achieved through sampling techniques.
Zero-G's first two vocal libraries will be released at the NAMM show in January 2004, followed by at least one more at the Frankfurt Music Messe in March 2004.
The system consists of a score editor, which does the scale, song-word, and expression processing; the Vocal Sound Generator, the engine that synthesizes the vocals; plus libraries produced by Zero-G and other soundware developers.
There's really no substitute for hearing it for yourself - sample audio tracks are available at the www.vocaloid.com site.