KAITO V3


 * This is an article about the KAITO V3 software for the VOCALOID3 engine.

History
''It is unknown when production switched from VOCALOID2 to VOCALOID3, however, it should be noted that even Crypton Future Media themselves did not know if this update would be for VOCALOID2 or VOCALOID3. Use of the VOCALOID2 version can be found in Project mirai''

KAITO APPEND: 2010
In October 2010 it was noted that CFM were considering updating Meiko and Kaito to VOCALOID2 or VOCALOID3.

Crypton announced in December 2010 that they were updating their old VOCALOID vocals. However, it was unknown if MEIKO and KAITO would be VOCALOID2 or VOCALOID3 as they were still in the planning stages.

Recording of Naoto Fuuga's voice started on December 10, 2010, beginning with test recordings. Several voicebanks of different expressions like the Appends were produced for the product.

On the 10th of December, tests were done on KAITO's voice. Wat confirmed they were discussing the marketing of the KAITO product. The project development was also confirmed in December to be MEIKO > Megurine Luka >> KAITO.

Testing was being carried out on a variety of tones such as a "Power" append that came from deep within the stomach. On the 17th December 2010, 4 appends were mentioned:
 * "Mellow"
 * "Serious"
 * "Sweet" - this was abandoned for the "Solid" Append.
 * "Solid"

Wat mentioned he would be asking for the thoughts of female producers on KAITO.

During the thoughts behind the update, it was noted that the differences between Miku and KAITO's low voice were ridiculously large and this was also reflected on the engine itself, leaving Wat wondering if they were still compatible. There was also the question of how to handle the male vocal, speculating if the "voice of the character" could be hung onto with a voice such as his. Wat discussed the problems with KAITO's voice that he had been presented with, stating since KAITO's vocal was a nice all-round voice, getting to act out the "expression of colour" previously seen in the Appends was challenging, as he felt the voice was already beautiful enough.

On the 25th of December, Wat announced that he had been charged with the sharing of information on Kaito on twitter. One of the concerns at the time of development was that KAITO's vocals they were working on were failing to meet quality expectations of consumers; more work was needed.

V3 KAITO: 2011
The first commercial usage of KAITO's Append was by yanagi and kaoling on the album "VOCALOID Minzoku Chou Kyokushuu (VOCALOID民族調曲集)", where he sings the song "Sennen no Dokusou Ka (kaoling mix)". The VOCALO APPEND album later used beta voicebanks of KAITO's "Normal", "Soft", and "Whisper" voicebanks singing "Lost Destination" with Kagamine Len Append.

In January, CFM noted that the silhouette has a transparent muffler. On February 17, 2011, Crypton released a temporary silhouette of KAITO's Append. They said that it has not been decided whether he will have a transparent muffler since there was previous outcry from fans.

In March, Wat mentioned work on KAITO's vocal had encountered a problem which halted progress. He did not know the release date yet.

In April, work on KAITO's voice resumed. Work on the vocal along with other projects had come to a halt. Wat mentioned wanting to do a vocal with a soft tone as well as a down toned voice. The voice of KAITO was going through checks on the α stage. On the 13th April, 2011, it was confirmed 6 vocal expressions had been recorded, two of which were dropped and the remaining were being brought forward. One had reached completed alpha stage already. The Character Vocal (Hatsune Miku, Kagamine Rin/Len and Megurine Luka) series Appends had been created from vocal performance by their voice providers. The new KAITO Appends were created by adding echo, force and tension to the samples.

Crypton Future media also noted that several voicebanks had been recorded with new high tech microphones through a series of trial and error instead of the standard U87ai, these were:
 * "Vivid"
 * "Solid"
 * "Power"

These vocals were done as a series of experiments with new equipment using Naoto Fūga's voice.

On the 11th of May 2011, Crypton went back to Yamaha to announce the current KAITO development, whose demos were at the time suspended. Crypton apologized for the slow development to their fanbase and enhancements to KAITO's Whisper Append were being made.

On the 28th of May 2011, demos of KAITO's Append "Whisper" and "Power", along with Miku's English voicebank, were posted on Nico Nico Douga.

On the 3rd of June 2011 the demo からくり時計と恋の話 was released using an alpha version of KAITO's "Soft" Append.

KAITO's "Soft" Append was considered by Crypton as KAITO's most natural vocal and was described as requiring little editing to sound good. "Soft" had also left α and entered β now. It was confirmed by Wat that release would be unknown for sometime, but would continue to give out information as soon as possible on the developments of the software. Wat noted during development that "soft" was being adjusted in relation to "whisper". "Whisper" was also still being adjusted, as well as other data which needed to be adjusted in a hurry. According to Crypton, lowering the gender factor on KAITO's "Soft" Append renders it sounding like a female vocalist. Although it was cute, it was considered a blunder on their part. So they tried to remove as much as they could.

By 11 June, Wat tweeted they were in the middle of the KAITO recordings and working hard for the first time in years to get the vocals completed. Wat noted they were trying to finish the main voice of KAITO to give them extra time to work on "Whisper". Later it was confirmed that by the end of the month, "Whisper" would be at the same stage as "Soft". Later they announced that the data had been settled on "whisperβ".

Wat reported that KAITO's "Whisper" vocal would have a demo due soon. Wat also referred to the process on KAITO English as "annoying". Also, in July, in relation to the introduction of Hatsune Miku in America, Wat noted that there was a high level of female Vocaloid fans in America compared to Japan. Based on this, Wat planned to test the demand on KAITO to see what the reaction was. Crypton also finished recording KAITO's vocals after work was previously suspended for an interval. Crypton was in discussion with Fuuga, KAITO's provider on the image and vocal outputs. A classic KAITO Vocaloid style recording had also been completed, one which will sound close to the vocal tone of the original voicebank that users were familiar with. It was then reported that there will be some wait for more information.

On October 14, 2011, Crypton updated via their twitter, stating that English KAITO will take some time to produce. Not long after, it was announced via Piapro and Miku's FB that a surprise concerning KAITO will be revealed at the 2011 New York Anime Festival. When October 16, 2011 came, the panel revealed a demo of KAITO's English VB. Footage of the demo was captured and uploaded onto YouTube:

It was later tweeted that KAITO's English voicebank was in its alpha stage, and that it still needed brushing up on. Checks were also done on the English vocal in December. KAITO's lower tones also had to be fixed.

Wat tweeted on the 1st December 2011, noting that all previous demos had originally been rendered with the VOCALOID2 engine, and things were being redone for the VOCALOID3 engine.

Crypton also ran a contest involving Kagamine Len Append, where the winner received exclusive information on KAITO's update.

V3 KAITO: 2012
In February Wat discussed the differences between the CV series (Hatsune Miku, Kagamine Rin/Len and Megurine Luka) and KAITO/MEIKO. He stated that their appends were being adjusted using editing techniques to achieve the results, whereas the CV series Appends had been done by acting performance. English KAITO's voicebank was reported to have received adjustments to a few of its diaphonetic data in February 2012.

Crypton Future Media began looking for Japanese producers who can use English capable VOCALOIDs to work with for the English version. KAITO himself is currently being compared to overseas male English VOCALOIDs. On 23 February it was announced that they were able to talk about the progress of both Japanese and English KAITO voicebanks in production. The company is exchanging VSQ files with producers to see how the current English VOCALOIDs compare with KAITO. Several overseas producers were now taking part with KAITO's production. KAITO's double vowel and vowel to vowel sounds had been adjusted after a report came back on their flaws.

It is confirmed that KAITO's clothes would not change in March 2012, they will just get a facelift. The boxart will also use a 2.5D style drawing. In early March 2012, Kenmochi Hideki was given the voicebank to make further adjustments. The English vocal was sent to Kenmochi Hideki to check the vocals. Crypton noted that all that was left was for KAITO's vowels to be adjusted.

In April, Crypton commented on "Whisper"'s use of triphones as "wonderful". KAITO's "Normal" voicebank was also officially declared as part of the new version. Wat also confirmed the manual will contain a history of VOCALOID KAITO.

In May, KAITO's voice was still being adjusted.

Checks were run on KAITO's voice in October. Further adjustments were made to KAITO and Miku in early October prior to their vocals demostration. Crypton were also considering what to call the default vocal for KAITO, they had suggested names like "Neutral", "Natural" and "Default".

An English demo of Miku and KAITO was shown NYCCon 2012, both were still in their beta stages. KAITO's V3 "Whisper" and V3 vocal were shown. KAITO's English vocal release was expected to be "by the end of this year or the beginning of the next".

In December it was reported that news on the update of KAITO is due within the next few days.

According to Crypton Future Media, there had been some delay on KAITO's V3 version due to their desire to include "Piapro Studio" with his package.

V3 KAITO: 2013
He was set for a February 2013 release.

In early January 2013, adjustments to KAITO's voicebanks were completed and KAITO was set to be released mid-February 2013 as scheduled. DTM Magazine is due to run a series of tutorials on how to use KAITO V3 for beginners. The idea of the bundles being offered with KAITO V3 software is to give users everything they need in basic sound making. The reason why KAITO has 4 vocal libraries is to minimize the amount of troubles they would encounter with the pack. A special discount was offered to VOCALOID1 KAITO users, the price based on the difference between VOCALOID1 KAITO and the CV series price.

KAITO V3 was released on February 15, 2013.

Mac Update
In late 2013 it was announced that Kaito would have a Mac release for the VOCALOID NEO engine. This version will be released for free with purchases of the Windows Kaito V3 product version. It was later announced that Kaito V3 Mac will be released on the 15th of February 2014.

Future Plans
In 2014 Wat mentioned playing with the wavelengths of experimental Vocals he hoped to one day release. These 3 vocals he mentioned were Kaito "Light", Miku falsetto and Meiko "Hard".

Demonstrations
Demo page

Recommended

 * Windows XP (32bit) / VISTA (32bit) / 7 (32, 64bit) ※ Windows 8 in (64bit) 32bit compatibility mode (WOW64): OS
 * Recommended Intel Core 2 Duo 1.8 GHz or more: CPU
 * Recommended 2GB or more: RAM memory
 * Free space of 25GB or more
 * Other: Display / Internet connection environment of video card / 1280x768px or more corresponding to the driver of the DVD-ROM drive / sound device / OpenGL 3.0 or more (when activated)

Voicebank Libraries
''KAITO V3 also includes Piapro Studio. ''