VOCALOID Trans-Pacific Project Unveiled

A new initiative titled “VOCALOID Trans-Pacific Project” (not to be confused with the more heinouscontroversial Trans-Pacific Partnership) was unveiled at DC Expo a few days ago. DC Expo (short for Digital Content Expo) is “an international event intended to shape the future of [the digital content] industry five or ten years into the future as well as promote the exchange of the latest information between participating researchers, creators and businesspeople active in this field.” According to a talk segment (recording available on Ustream) involving INTERNET, Bplats and Yamaha, the project aims to spread the creative culture behind VOCALOID to the US and eventually throughout Asia. INTERNET Co. Ltd. also unveiled a tentative release date of February 2013 for the English version of megpoid. This is the third year VOCALOID was featured at DC Expo, with simple demonstrations of the software and LiveAR performances held previously.


At the venue, the projector first played back related footage from previous years of DC Expo, including shots of HRP-4C MIMU as well as a LiveAR performance featuring Akikoloid-chan and GUMI dancing to Aoi Konbini de Aimashou and Megane no Mukou no Sekai. The video concluded with a series of on screen messages: “2012”, “VOCALOID meets Cool Japan” and “VOCALOID Trance[sic]-Pacific Project”. After the video showing, four panelists walked on stage, apologizing for being just a group of old men.

After everyone sat down, the panel chair, Masaru Ishikawa, talked briefly about what was shown on screen. He also mentioned that this was the third year VOCALOID was featured at DC Expo and a new project will be unveiled. Ishikawa then invited the rest of the panel to introduce themselves: Noboru Murakami from INTERNET, Takuma Miyazaki from Bplats and Osamu Ohshima from Yamaha. Finally, Ishikawa introduced himself as a researcher from University of Tokyo working in the Information and Robot Technology Research Initiative.

With introductions finished, Ishikawa summarized the aim and content of this talk show. This Trans-Pacific Project is essentially a collaboration between VOCALOID and the Cool Japan Project, the latter aiming to spread culture based on Japanese creative works as well as related industries throughout the world. The title “Trans-Pacific” was intended to convey the idea of crossing the Pacific and spreading VOCALOID culture over to the US, as well as to other Asian countries over time.

Bplats’s Yamazaki noted that there actually hasn’t been a VOCALOID product native to North America yet. Yamaha’s Ohshima then clarified that although there are several English VOCALOID sound banks, none of them originated in North America. Having one come out of the US would require a suitable business partner.


Before going further, the panel wanted to describe in more detail what VOCALOID is. Ohshima stated that while VOCALOID is simply vocal synthesis technology to Yamaha, people began writing songs with VOCALOID and others started drawing artwork, resulting in VOCALOID spreading far and wide.

As far as business was concerned, the biggest role Yamaha plays is the development of the technology and its licensing to business partners. One such licensee is INTERNET, which creates the voice banks and packages the two together into a product. Meanwhile, companies like Bplats operate in the background and handle the logistics of eventually setting up a product on a store shelf.

However, the panelists stated that VOCALOID embodied more than simply the business aspect. In fact, a “VOCALOID culture” as developed. One example given was a specific LiveAR showing outside a train station in Japan. Footage on the projector showed animated computer generated models of Tone Rion, Akikoloid-chan and GUMI dancing with the crowd, while Cyber Thunder Cider played in the background. This event was held outside of Seibushinjuku Station, and the original camera footage was combined with the 3D models using augmented reality techniques and then displayed on giant YUNICA VISION screens. Some of the people in the crowd didn’t really know what was going on, but a lot of them were happily partying along. Ohshima mentioned that perhaps this could be expanded into a digital signage business and also stressed that the music was created by “normal” creators. In a way, he said, this was like a guerrilla concert. Ishikawa noted that while people of his generation probably don’t have a very good idea of VOCALOID, younger generations, especially those in their teens, are very familiar with VOCALOID; VOCALOID songs consistently score very high on song popularity rankings for karaoke.

VOCALOID through INTERNET, Bplats and Yamaha

With an overview of VOCALOID and its culture out of the way, the panelists wanted to then go into more detail about their respective businesses. Murakami started by going over the release history of his company’s VOCALOID products, starting with the VOCALOID2 releases. He noted that although he started with sound banks recorded by singers, for some reason they also decided to release Gachapoid. Murakami then moved onto giving an overview of INTERNET’s VOCALOID3 releases, at which point one of the panelists wondered if there were perhaps multiple copies of him, since he worked personally on all these databanks. The slides then ended up with a shot of a white product box, stating that megpoid English had a tentative February 2013 release date. Apparently this information hadn’t even been sent out via press releases yet, so this was considered an “unofficial announcement”. According to Murakami, the recording for this sound bank had already been finished, and the panelists started discussing the differences between Japanese and English sound banks.

Apparently, while Japanese sound banks have approximately 500 or so phonemes, English has approximately 5 times the amount. However, unfortunately while the work is greatly magnified, they obviously couldn’t raise the price and still expect it to sell, and thus one panelist quipped that Murakami had to take a pay cut (in terms of income per phoneme). According to him, it also takes approximately three to four times as long to record the audio for the English sound bank.

Murakami then moved on to promote his company’s upcoming Singer Songwriter (SSW) 10 product, with a scheduled release date of November 9. He noted that although VOCALOID can provide the vocals, songs still needed backing instrumentation, which is where digital audio workstation (DAW) software such as SSW came in. Although VOCALOID3 stopped supporting ReWire (and thus direct synchronization with DAW software), SSW 10 claims to be able to synch with VOCALOID through the use of a VST plugin that VOCALOID3 Editor would load. This plugin sounds very much like V3Sync, a third-party plugin which let VOCALOID3 Editor talk with other DAW packages.

After Murakami’s overview of INTERNET, Ohshima proceeded to give an overview of the recently-released VocaListener, touted as a tool which can be used to copy the singing style of a vocalist from a recording. As an example, he showed GUMI copying Kaori Mochida’s performance of Every Little Thing, which was originally to demonstrate MIMU, but the MIMU voice bank isn’t publicly available. He noted that the provided vocals unfortunately had some reverb and thus VocaListener got confused by it, making the result kind of strange. VocaListener’s distribution is handled by Bplats.

And with that segue, Yamazaki went over what Bplats was about. Bplats provides many services, including cloud service brokerage and conversion assistance. However, Bplats also handles distribution of VOCALOID products. The company was involved with the CloudVST project (which enabled usage of VST plugins through the cloud) as well as the VOCALOID STORE, VOCALOID CHINA and the distribution of VOCALOID3 products. Recently, Bplats started a new label called VOCALOID MUSIC PUBLISHING (VMP), which aims to help VOCALOID artists distribute their music. Although not mentioned during the panel, Bplats was involved in the distribution of galaco and encouraged downloaders to sign up with VMP. VMP then sent out emails recently about their service, which submitted songs to various digital distribution platforms for a 50% cut of the revenue (after the distribution platforms take their cut).

VOCALOID Trans-Pacific Project – Slide 1

At this point the panelists realized they may have dwelt too long on preliminaries and moved on to talk more about the Trans-Pacific Project. A giant word-packed slide was shown on the screen detailing the project.

This first slide talked about one aspect of the VOCALOID Trans-Pacific project — setting up a foundation for a VOCALOID-using music business in America. The slide gives a background on VOCALOID as well as the lack of a basis for VOCALOID music in America, which is where the project would come in. The project aims to break down previous bottlenecks to establish the basis for a “very profitable” [VOCALOID music] business outside Japan. This will be accomplished through digital distribution of music, sales through both online and retail channels, promotions during events attended by music creators, construction of a foundation for interchange through VOCALOID as well as business-to-business marketing, all timed to coincide with the release of English sound banks for VOCALOID. These measures will be enacted in America, with its high Asian population and being the epicenter of a music business with global reach. The fruits of this endeavor would then be circulated through Asian cultural networks in the Asia Pacific region. The actual plan involves the following specific actions:

  1. Promotion of the sale of English VOCALOID
  2. Distribution of VOCALOID music through social media
  3. Promotion at New People in San Francisco
  4. Marketing and promotion at events in the US attended by music creators
  5. Partnership with the English version of niconico, niconico.com [though that has technically merged with niconico proper already]
  6. Promotion in collaboration with Japanese businesses

The main parties (or “producers”) involved with bringing this to fruition are: Yuki Seto (producer from Yamaha’s yamaha+ Promotion House) as overall lead, Noboru Murakami (CEO of INTERNET Co., Ltd.) to handle VOCALOID software, Seiji Horibuchi (CEO of NEW PEOPLE Inc.) to handle relations with the American side, and Masaru Ishikawa (CEO of think.communications and Specially Appointed Researcher at Tokyo University) to handle marketing and promotion.

The project’s primary schedule consists of the following:

  • From October 2012 to January 2013: content and website creation
  • From January 2013 to March 2013: Promotion both at retail outlets and through the media
  • From January 2013 to March 2013: Promotion for the sales of English VOCALOID
  • From January 2013 to March 2013: Promotion and marketing at creator-oriented events [e.g. conventions/conferences]

Lastly, the main slide provided a list of participating companies. They are: Bplats, Inc., Yamaha Corporation, UGC Publishing, Inc., INTERNET Co., Ltd., AHS Co. Ltd., 1st PLACE Co.,Ltd., STUDIO DEEN CO.,Ltd. and DWANGO Co.,Ltd.

VOCALOID Trans-Pacific Project – Slide 2

A second slide illustrates a second part of the project, with ideas aimed at creating a foundation for businesses rooted in VOCALOID user-generated content. In other words, the planners wish to elevate VOCALOID as a genre to a level commensurate with genres well known all over the world, like rock and fusion. They break this aim up into five smaller goals:

  • Increase inbound Japanese pop-culture tourism traffic to Akihabara and Harajuku.
  • Increase the media content market in America via English VOCALOID.
  • Increase opportunities for talented young Japanese creators to debut in America.
  • Ripple effects affecting distribution and other businesses.
  • Ripple effects emanating towards the Pacific rim, including countries such as China, Indonesia, Thailand, Malaysia, Singapore and Australia.

They also detail several more specific plans:

  1. Promotion timed to coincide with release of an English VOCALOID [with a picture of GUMI surrounded by glowsticks].
  2. Distribution of VOCALOID music — Broadcast of music by Japanese creators through internet streaming sites [with a flipped picture of YouTube].
  3. Sales through brick and mortar outlets — sales of physical product through NEW PEOPLE in San Francisco, a beachhead for J-Pop.
  4. Promotion of the project at creator-aimed events, including International CES and the NAMM Show.
  5. Constructing a foundation for VOCALOID interchange in America — collaboration with the English version of niconico on niconico.com.
  6. Business to business marketing in America.

This project will involve the participation of both VOCALOID-related and unrelated businesses, such as Bplats, Yamaha, INTERNET, AHS, 1st PLACE, STUDIO DEEN and niconico.

VOCALOID Trans-Pacific Project – Discussion

While the slides were running, the panelists commented on them, mentioning January events in Las Vegas (CES) and Anaheim (NAMM). Murakami stated that his biggest goal was for American creators to use megpoid English. Yamazaki noted that Bplats will be providing the distribution infrastructure required. He also played a curious clip of a demo song sung in three different languages. SeeU sang the Japanese and Korean versions of Kimpaksa’s “Alone” (lyrics by 45 degrees), with Luo Tianyi covering the Chinese version of the song. The panelists agreed that usually, there are two types of people who enjoy VOCALOID — those who like it in Japanese and those who’d like to experience it in their own native language, and they note that America is likely like that as well. Ohshima commented that Yamaha is known internationally for their musical instruments and they would also like for VOCALOID creativity to bloom as well.

Finally, Ishikawa added that he is happy that VOCALOID is getting more popular in America. However, he wants the idea of VOCALOID as a musical instrument to spread more and want users to make new media using VOCALOID and spread them on content sharing websites, getting fans to comment on them, etc. Without this cycle, he feels that the VOCALOID phenomenon in Japan wouldn’t be able to be replicated elsewhere. He would like the business process of nurturing VOCALOID in Japan to expand step by step into America.

And with about 20 seconds left on the clock, the panelists said the closings to conclude the panel.

24 thoughts on “VOCALOID Trans-Pacific Project Unveiled”

  1. Well, it looks like I’ll be looking forward to seeing plenty of fun-filled Vocaloid events when it’s next year. 🙂
    Speaking of that, I didn’t know Rion Tone was featured in Gumi and Akikoroid’s Cyber Thunder Cider concert. 😐

  2. “heinous” ? As much as I appreciate your contributions, I think it’s really inappropriate to use them to prop political opinions that are not related to Vocaloid. There is a lot of good and bad about the TPP that can be discussed but I think Vism is really not the place for that.

    1. I take it that you do not appreciate hyperbolic jests 😉 Though I suppose if you seriously feel that I am being serious with that comment, I can change it to something more informative to avoid misunderstandings…

  3. That was one great and informative article, thanks for that 🙂

    I am very surprised by the explicit intent to establish Vocaloid as a full musical genre in itself. That is far from the initial goal of Yamaha to provide a replacement vocalist. But admittedly many, if not most of the composers are now using Vocaloid in their own creative way, though it will be hard to classify a lot of composers just using better and better voicebanks without the intrinsic roboticness the V2 / V1 engines had.

    Since Gumi will presumably be released simultaneously to Miku, I wonder why CFM is not collaborating with this initiative… That said, they’re probably going to use Miku as their salesperson while InternetCo is (rightly IMO) going to work on the collaborative ecosystem first without pushing Gumi too much. But ultimately I hope that both ways of trying to promote Vocaloid in the West will compliment each over.

    Speaking about Miku, InternetCo will have delivered an English Voicebank in somewhat one third of the time it took CFM… I wonder why and just hope Gumi won’t be of lower quality. But Maybe Megumi Nakajima is a fluent English speaker ?

    I hope that the other western Vocaloid companies will also benefit from this project (maybe it is already expected by their “ripple effect for other businesses”).

    1. I think the reason whyInternet produced a englsih voicebank quicker is just because that CFM have taken a lot of time with this for several reasons.

      1) They have several projects going at once; Luka append, V3 versions of of all 5 Vocaloids, Kaito and Meiko have new vocals in their updates, and each vocal has a english voicebank. CFM likes doing bulk developements, then releasing things s part of the plan, so you get pretty much everything happen at once. As a consquence though to this bulk software production, all new vocals have been confirmed as “not happening until its over”. Meaning no Project If or that male vocal we’ve been promised. On top of this, original Meiko and Kaito were going to be a vocaloid 2 release, but this didnt’ happen. Thy pretty much had to restart from scratch in terms of building up the database, and reconfigure everything so it works better with Vocaloid 3.

      2) Recently, CFM moved buildings to a new home, they are now a bigger company.

      3) Unlike other studios, everything is done “in-house” so CFM have full control over everything. I don’t know if Internet co., Ltd are the same, but CFM seem to be capable of producing a vocal for sale in 2 months, but send a lot ofediting to get things how they want.

      This is without considering Yamaha licensing and so forth.

  4. Crypton out of the picture; good thing or bad thing?

    For me:

    1. I don’t care that much then (no Miku)

    2. Which is fine since I’m perfectly comfortable (rather, actually) with VOCALOID staying in Asia.

    Although it would be interesting to see what happens, I guess.

    1. Crypton intends to get English Miku out soon (they’re aiming for next Spring), and hopefully if she does well we’ll get the others. They really do need to pull their heads out of their asses, however, as the demand for their products is huge but they seem to be putzing around.

    2. Why does everyone forget that VOCALOID was originally in English with Leon and Lola, and still is very good with English? It was never exclusive to Asia in the first place!

  5. I like the initiative for this project, but I really hope that a real english vocaloid would be released at this time. Hopefully Avanna would be released around that time, since I don’t think Gumi english is going to impress an audience who is not familiar with vocaloid. If anything, it might just push away potential customers with the strong accent since it woule be “a japanese thing”. (Same applies for Miku english as well, unless CFM starts to put her on billboards or something since Miku has a slight advantage of being well known, but even then, that can easily backfire on CFM also.)

    1. Wait, you’ve heard English GUMI demos? I wasn’t aware those were available yet. Can you link to them? And how strong was the accent compared to the avalable English Miku demos?

      1. GUMI’s voice provider, Megumi Nakajima, seems to be able to sing English quite well! Check out her song “Flower in Green” which is all in English, for example…

  6. This seems very similar to the 2004 NAMM show demonstrations of LE♂N and L♀LA and that ended pretty poorly once the hype died down and people got a good listen to how bad english sounds in Vocaloid. Something tells me that it will be a nice attempt but I doubt most musicians will actually use Vocaloid thinking it is just “that Japanese thing”. I will never really understand why Japan suddenly wants to cater to the English side of the community we are known for usually screwing things up so handed us the keys to the program is kind of like giving a loaded gun to a person who will shoot you. I personally think think that English KAITO, English Miku, English Gumi, and Avanna are gonna do very poorly since they are gonna show off how poor the V3 engine is with English compared to Spanish, Korean, and Japanese. (Germanic based languages are just not very musically compared to Asian and Latin based languages)

    TL; DR Probably gonna flop.

    1. ^haha that guy said what I didn’t feel like saying and more
      To answer,

      It’s a logical next step for those who have now a tasted success.

    2. This guy is alway anti-engloid, every time its mentioned he shows up. He is a troll.

      I won’t say english Vocaloid will be a flop, but its not going to produce the same impact as in Japan. This is not a surprise because outside of Japan, Vocaloids are generally having mixed results. The Japanese culture is different to even others in asia, so things that do well in Japan don’t always do well elsewhere. :-/

      I mean, its been confirmed SeeU failed to met expectiations of her producers and the Spanish fandom is still quite far behind most other fandoms. The English vocaloids don’t sound as bad as you make them out to be and quite honestly, there are bad songs out there for Japanese Vocaloids.
      Now the English speaking part of the song industry is accountable for a lot of the music industry and is one of the most successful. English is spoken by a lot of people as a second language, again, this is something you have to consider because English songs can go well even other countries.

      1. I would be kinder to English Vocaloids but each time we get a new one they just keep going down in quality. There recording methods are still the same since 2004 still doing CV-VV-VC and The timbre quality is just poor and there are many many glitches one the english side. (Like Tonio and Sonika @ glitch, Prima activation error, Olivers and Sonika background hiss, and incorrect programing of phonemes (Seriously the word “Of” should not be programed “V v” it should be programed as “O: v” since that was literally one of the examples in the X-SAMPA programing guide)) Meanwhile the Japanese side has tried new styles of recording like VCV and multi-tone and also has attempted to fix up banks that had issues with releasing updates. Most of the english vocaloids can’t even pronounce the basics of the english language because the voice banks and engine can’t figure it out or will slur it together to become a jumbled mess. (Which yes I know the Japanese ones did that as well but it is much easier to fix with a simple velocity change then it is with english. ( especially with @r phoneme; that is always disaster since it can’t be broken apart without the program messing up the Opening setting on @) That is why I am very negative towards engloids because they are just overall poor quality. You can give a billion examples of bad Japanese vocaloid song but compared to english vocaloid songs they are waaaayyy better.

        1. Actually, they haven’t been getting wo4rst in quality and the methods behind making them HAVE changed. Hence why you get classified as a troll, you only post to get a knee-jerk reaction from people and you BARELY know what your talking about. The same background hiss you conplain about in Oliver and Sonika is in Yuki, Miku Soft, Ia,Gumi.

          Also, Ia slurs her words, so do several other vocalbanks in Japanese. Slurring is a common problem with vocaloids, its a issue with the engine not the langauge and often comes out on faster pace songs. Basically, its a result of bad synchronisation of lyics and tempo.

          Youer negative towards Engloids because your looking to troll everyone who lkes them. I know a lot of bad songs for Japanese Vocaloids, anything by that miki-guys is pure crap (the guy who posts clapped together VSQ using miki). Also, Miku, the Kagamines, Lily… Their missing sounds, Len’s act 1 can barely do Japanese.

          Now, if you want to be negative its fine, if you want to troll its not, buyt seriously, your boring everyone, we’ve already proven your bias and stupid in the past, why don’t you levae everyone alone? No one likes a troll. Even if your not a troll, your too blind to the issues with the Japanese Vocaloids to be bothered with even when their flaws can put them at WORST then the Engloids (believe it or not).

          Also, take note, Spanish does stress some words, in fact it has severakl things in common with the english language (though note as common with german and French). Also, Jpnese still uses the Xampa language and when you consider most of the progress for Japanese came late in 2010.

          1. Indeed that hiss is in those banks that you mentioned but Both Miku and Gumi have had new banks made that have removed it while Zero-G flat out stated they won’t fix it even though it has been brought to there attention. ( I’ll give you Yuki since most have mentioned it and AHS software seems to be in the dark on what they are going to do with her and the other AHS Vocaloids)

            Miku, Aoki,Len, Rin, and Lilly also are noted for having slurring but with the Japanese bank it is less noticeable to where it does not through off the entire meaning of a word and it can be fixed with a simple velocity and dynamic change (Or in Len and Rins case a DAW cleaning) while with the english vocaloids it because it is in the recording itself along with the engine making it unfix able no matter how much time and effort you put into it. (MIRIAM, Prima, and Sonika are perfect examples of this)

            As for the Japanese Vocaloids missing Phonemes that is very true but English vocaloids are missing even more of them. For instance there is no recording of the phoneme 3 (ɜ), 4 (ɾ), 5 (ɫ), 1 (ɨ), and a dozen other needed recordings ( Like rising tone,dental clicks, and nasal release just to name a few) which they tried substitute with other phonemes that just don’t work.

            Also on top of that the dictionary is no where near completed. Worlds like ” euphoria”, “wretch”, “wind”, and “tear” are either not even in the program or incorrectly programed. Even when brought up at discussion panels about improving the dictionary the usually response is ” Have the fans figure it out” which basically shows that english companies knew that these problems existed but just didn’t really care and released the program uncompleted. You could make this same argument with the Japanese vocaloids but the issue is that all of the phonemes are in the Japanese libraries while the english ones do not making it so some of the words are not possible to do thus bringing them down in quality.

            As for Spanish yes it does use stress in its language because it is a stressed based language like english. However the Spanish vocaloids are much better then the english because they indicate where a stress will be instead of in english where it just shows up where ever. If I had to put the vocaloids in terms of quality it would probably go somewhere around this Japanese, Korean, Spanish, Chinese, then English. (Of course since SeeU is the only Korean vocaloid Lou is the only Chinese Vocaloid it is kind of difficult to judge) Also if UTAU were allowed in that rank I would put UTAU one above english since even the CVVC banks in UTAU are better quality then most English Vocaloids though not by much.

            As for me going away, Thanks but I think I will stay. Pointing out how flawed the program is for english is to let potential users know that it really isn’t worth wasting the money for the english banks and that they should just focus on the Japanese banks. Maybe if the companies would think outside of the box with the english banks and do proper recording then other vocaloid fans and myself would stop being so harsh on them. Sure the Japanese side has made lots of mistakes and has lots of bad songs but unlike on the english side they go back and try to fix them.

    3. You have to remember that Lola and Leon were a much weaker engine. Vocaloid in general has come a long way, and with the large popularity of Vocaloids as well as the upgraded engine, I think this stands to have decent potential.

      Remember, a lot of the demos we’ve seen for Vocaloid3 English banks were not only incomplete, but also tuned by non-native speaker and made very quickly. I think that, in the hands of an English speaker with lots of time on their hands, we could see some cool stuff. I know I intend to buy Megpoid when it comes out, and whenever Crypton gets off their ass and puts out their English Vocaloids (they said maybe next Spring for Miku) I’ll buy those too.

      1. I believe you mean “On a much weaker engine” which one could argue that isn’t exactly true because the V1 engine gave you much more options to work while V2 and V3 were more user friendly by removing the extra options. (Resonance being the major player) As for Vocaloid coming a long way it has for the Japanese side but not so much for the english side.

        I’ll give you that a lot of the demos were done by non-english speakers but the way they have gone about it as in purposely giving a heavy accent, making the product sing out of its intended range, or using Luka to fill up parts of the bank that were not recorded kind of defeats the whole point of an english demo and even an English bank.

  7. Certainly interesting. Vocaloid’s mixed success in America may be some problems but they currently have a growing fanbase, so it may be a good idea to make it more connected to the American audience. I certainly hope for more English-speaking vocaloids, as their technology is always improving. All in all, a very informative article.

Leave a Reply

Your email address will not be published. Required fields are marked *