Vocaloid Software

New Yamaha VOCALOID Singing Synthesis Software Generates Superb Vocals On A PC
Yamaha Corporation has developed new software that allows song writers to generate authentic-sounding singing on their PCs by simply inputting the words and notes of their compositions.
Named VOCALOID, the software synthesizes the sound from “vocal libraries” of recordings of actual singers, retaining the vocal qualities of the original singing voices to reproduce real-sounding vocals.
“The software also features simple commands enabling users to add expressive effects,” says Yamaha Corp Managing Director and Chief Technical Officer Masatada Wachi, “and as VOCALOID runs on Windows-based PCs, we hope that amateur enthusiasts as well as professionals will enjoy creating music with great-sounding vocals.”
Yamaha will unveil and exhibit its new VOCALOID software at the Musikmesse slated for March 5-9 in Frankfurt am Main, Germany, and the 114th Audio Engineering Society (AES) Convention scheduled for March 22-25 in Amsterdam, the Netherlands.
VOCALOID vocal-synthesizing software overcomes a major hurdle composers have faced until now due to the limitations that technology has placed on their ability to freely create songs incorporating singing.
The development of a full complement of MIDI peripherals–notably those that are XG format compatible–permits the production of high-quality music on a PC or other low-priced equipment, and many music industry professionals are producing CDs of music created with just such computer systems.
Problems have arisen for composers, however, because existing technology prevented them from achieving the same success when their compositions involved vocals. Though systems and software that synthesize singing voices exist, the work entailed in making them sound real was complicated and onerous. Even after much fine-tuning, synthesizing vocals to be indistinguishable to listeners from real singing was still virtually impossible.
The popularization of sound samplers containing a wide range of prerecorded sounds has led some producers to try creating music with vocals using samples of prerecorded human-voice snippets. But since this method’s range of words and vocal expressions is very limited, it too leaves much to be desired.
Yamaha’s newly-developed vocal-synthesizing VOCALOID software changes all this.
By just inputting the melody and words on their PCs, users can produce the vocal parts for their pieces with no further work. The synthesized sound retains the vocal qualities of the original singers’ voices because VOCALOID synthesizes the sound from “vocal libraries” of recordings of real people singing. The software also features simple commands that allow users to add expressive effects–such as vibrato and pitch bends–to their synthesized vocals.
Further development of the range of available vocal libraries will make the production of songs using vocals in a wide range of voice qualities possible using VOCALOID. Currently, VOCALOID can generate singing in Japanese and English.
Yamaha intends to concentrate on licensing VOCALOID to selected producers of vocal libraries and software marketers. The company has already agreed on the basic framework of a license agreement with a number of sampling-CD development and marketing firms, including Crypton Future Media, Inc. in Japan; Best Service in Germany; and Zero-G Limited in the UK.
Yamaha hopes to commercialize VOCALOID by the end of 2003 or early in 2004.
VOCALOID uses Frequency-Domain Singing Articulation Splicing and Shaping, a vocal (singing-voice) synthesizing system developed by Yamaha.
With this system, the “singing articulations” (collections of voice snippets, such as of phrases, and snippets of vocal expression variations like vibrato) needed to reproduce vocals are collected from custom-produced recordings of accomplished singers and put into a database after conversion into frequency domains.
To synthesize vocal parts, the system retrieves data consisting of voice snippets, applies pitch conversion, and splices and shapes them to form the words of a song as input by the user. As this processing is done at the frequency-domain level, pitch can be easily changed according to the specified melody, and the voice snippets can be spliced in a way that reproduces smooth-flowing words.
VOCALOID itself consists of a score editor, which does the scale, song-word, and expression processing; the Vocal Sound Generator, the engine that synthesizes the vocals; and libraries (each comprised of a pronunciation database and a timbre database) for each vocal. New vocal libraries can be created by recording real voices pronouncing basic vocabulary and reproducing variation effects (such as vibrato) according to templates.