Synthesizer V OSCAR: New Masculine Mandarin Voice from VOICEMITH

For a while now, there’s been hopes that the rest of VOICEMITH’s voice synthesizer cast (OSCAR (male UTAU), JOAN (female UTAU), and ZETA (female DeepVocal)) would be joining XIA YU YAO in the Synthesizer V production. After her crowdfund was launched, when asked if the others would be coming along, YU YAO’s Twitter only had this to say:

After YU YAO released, there weren’t any announcements about the website’s adjustments, but the navigation had changed to have YU YAO. The rest of the cast was removed. Each character had have their own webpages and were not available through the navigation, but if you still had the links, you can still access them.

On March 20, OSCAR got official Twitter and YouTube accounts. The accounts used his “neo” design from 2019 (illustrated by Shenzi (神子)), which was used as a sort of VTuber-like marketing tactic where the voice provider (Myers; 邁爾斯) would speak and sing as OSCAR (however, such activities ceased sometime after). The handle of the Twitter account is “VM_OscarAI” which lots of people were curious about, however, it was not confirmed what synth the AI voice library would be made for.

If you’re curious as to how Myers sounds, here’s the two neo videos he did. The first his him speaking as OSCAR and the second is him singing in English with the VTuber LanCee (another VOICEMITH cast member).

On April 1, best known as April Fool’s Day, OSCAR’s accounts posted a new video that mentioned “OSCAR AI”. As one would expect, it was a joke, but the last part of the video confirmed that a crowdfunding campaign for Synthesizer V AI will be expected and to stay tuned for more information in the future!

Twitter user @/andresfelipeb25 brought up the OSCAR neo performance with LanCee, noting that Myers can sing in English, hoping that OSCAR can be trained with English songs. The response? VOICEMITH will try to have the training data in Mandarin, English, and Japanese. Fingers crossed that happens!

Lastly, it’s been confirmed that OSCAR AI was already in the recording stage. The confirmation of the engine was confirmed in the April Fool’s video, but here’s another tweet to further support it.

Curious about OSCAR’s history as an UTAU? Check him out on UTAU Wiki!

What are your thoughts? Are you excited that possibly more of the VOICEMITH vocals might be able to make it to Synthesizer V? What are your hopes for OSCAR and what do you think of him so far?

「いいね!」 5

OSCAR AI crowdfund and a XIA YUYAO crowdfund are ready to go! Details to release later on perhaps?

https://twitter.com/vm_oscarai/status/1686268126286241792?s=46&t=Uf7yaGuTpN8LpgYGv7mbgg

https://twitter.com/YuyaoOfficial/status/1686274372011782144

「いいね!」 2

OSCAR’s crowdfund is up now, along with XIA YU YAO’s upcoming E.P! Since this is a thread that’s about OSCAR, I’ll only be focusing on his information.

Key takeaways:

  • The fundraiser will end on September 18 (September 17 at 23:59)
  • Physical rewards, including OSCAR’s physical box (which will have a download card and NOT a disc) will be delivered in mid-October
  • OSCAR’s digital download is expected for early December, marking this as his final release.
  • As a correction, OSCAR’s voice provider’s name is officially written as “Myles”, not “Myers”. Initially, we only knew the phonetic writing of his name. (EDIT: Upon inspecting the FAQ, he is also referred to as “Myers” still… so…)
  • Introducing GENE as the illustrator for OSCAR’s physical box. Jinko remains as his standing artist and this illustration is still used for the digital icon.
  • 4 vocal modes are to be expected: Power, Soft, Classic (noted to be good for older songs), and Nasal. Very self-explanatory modes.
  • 3 more vocal modes may be produced if the crowdfund raised 900,000 NTD. These would be known as “Rock”, “Special” (Chinese opera styled singing), and “Who” (which is I guess considered an unclassifiable type of voice)

If VOICEMITH makes mini updates and I’m still able to edit my post, I’ll make revisions while I can. The forums won’t allow me to make too many posts directly after another within a small amount of time. Also for Twitter links only, since embeds are broken, I’ll do my best to provide images along with the link.

1st update:
Paypal support was added. If you use their direct link, you can pledge for a digital copy of OSCAR.

Source: https://twitter.com/VM_OscarAI/status/1686657907612643328

2nd update:
For those who previously supported XIA YUYAO’s Synthesizer V AI library crowdfund, 400NTD coupon codes were distributed to their emails! Be sure to take advantage of those if you’re interested in the current crowdfund!

Source: https://twitter.com/YuyaoOfficial/status/1686662751228493824

「いいね!」 4

A preview of Myles singing “The Blower’s Daughter” by Damien Rice was posted to give people an idea of how OSCAR may sound in English!

https://twitter.com/VM_OscarAI/status/1688409783693631489

Also a new reward was added! It’s a pack with OSCAR’s physical box + his merchandise!


Source: https://twitter.com/VM_OscarAI/status/1688162869320536064


EDIT 1: The full illustration for OSCAR’s physical boxart was revealed!


Source: https://twitter.com/VM_OscarAI/status/1688436368773672961

EDIT 2: It was clarified that GENE’s illustration is the outer sleeve of the box. When you take the sleeve off, it reveals the standing art by Jinko.


Source: https://twitter.com/VM_OscarAI/status/1688805852650848256

A new singing example of Myles singing “All of Me” by John Legend was shown.
https://twitter.com/VM_OscarAI/status/1688730504449982464

EDIT 3: Example of Myles singing “Shut Up and Dance” by Walk the Moon.
https://twitter.com/VM_OscarAI/status/1689090354329583616

EDIT 4: They just started showing examples of Myles singing in Mandarin. Here’s “雨停”, his very own original song.
https://twitter.com/VM_OscarAI/status/1689477873323950081

EDIT 5: The team headed back to the studio for a second time to record samples for OSCAR’s voice database!



Source: https://twitter.com/VM_OscarAI/status/1689878492673081344

「いいね!」 2

anyone have discount code?
Edit: found it already …

Some new Twitter updates!

Sample of Myles singing “Lost Stars” by Adam Levine
https://twitter.com/VM_OscarAI/status/1690952207955574784

The coupon code is set to expire on August 20, so use it as soon as possible before it’s no longer valid!


https://twitter.com/VM_OscarAI/status/1691635642898518427

And in celebration of reaching 150,000 NTD, computer and phone wallpapers by LLG, Jinko, and GENE are being given out.


https://twitter.com/VM_OscarAI/status/1691644542452457707

Still need to reach 800,000 NTD minimum and 900,000 NTD for the extra Vocal Modes! Keep it up and going!

EDIT 1:

A clip of Myers’ recording for OSCAR!
https://twitter.com/VM_OscarAI/status/1692776527308714060

And not long ago, the crowdfund has over 100 backers! Still got less than a month left to try to fund for the project and its milestone for the extra Vocal Modes!!!


Source: https://twitter.com/VM_OscarAI/status/1693456095422140826

「いいね!」 3

So it seems that Voicemith has changed the way that OSCAR’s Vocal Modes would be distributed.

By default, OSCAR will come with 3 Vocal Modes: Power, Soft, and Rock

If he reaches 900,000 NTD, 2 more will be added for all users:

For those who supported his crowdfund, they will have exclusive access to 2 Vocal Modes: Nasal and Special. Please note that these would NOT be available for regular release.

So depending on how the crowdfund goes and who supported the crowdfund, the variations will turn out like this:

If OSCAR does NOT reach the milestone: Regular users will have 3 Vocal Modes (Power, Soft, Rock), Supporters will have 5 (3 mentioned + Nasal and Special)
If OSCAR reaches the milestone: Regular users will have 5 Vocal Modes (Power, Soft, Rock, Classic, Who), Supporters will have 7 (5 mentioned + Nasal and Special)

Source: https://twitter.com/VM_OscarAI/status/1694637057757552878

「いいね!」 2

Made my pledge this morning, thanks.

「いいね!」 1

OSCAR is asking for feedback in regards to this change.

HI! If we were to make the exclusive Nasal and Special vocal mode from the fundraising campaign available as DLC in the future, giving those who need them the option to make additional purchases. Welcome to let us know your comments.

Source: https://twitter.com/VM_OscarAI/status/1696764241527931013

「いいね!」 1

Crowdfund goal has been lowered

Thank you for your support. We truly hope that this fundraising campaign can be successful. Therefore, we have lowered the funding goal to NTD$550,000. With nine days remaining, to make this fundraising campaign a success. thank you all!

Source: https://twitter.com/VM_OscarAI/status/1700039546178875455?s=20

「いいね!」 2

Just to make sure, is the below what one should pick to back the project? And if so, why does it say 2,500$?

「いいね!」 1

This is his digital copy

Please keep in mind that this is in NTD (New Taiwan Dollar) and not USD (United States Dollar).

「いいね!」 1

With the exchange rates it ends up being about the same price you pay for soundbanks at other places.

they have discount code, you can PM them thru twitter

OSCAR’s crowdfund had reached the main goal~!

We’ve successfully reached and exceeded our fundraising goal, surpassing NT$550,000! We’re truly grateful for everyone’s contributions and support along the way. It’s been a delight to celebrate reaching our fundraising target with all of you today. Thank you!:heart:

Source: https://twitter.com/YuyaoOfficial/status/1702642837505130782

「いいね!」 3

yesss … I am so stoked

I’m a bit late, but this is Oscar’s first demonstration, showing off his vocal modes.

From the video description (translated with DeepL):

  1. There are two versions of the library: Standard & Deluxe.
  2. Standard version contains 5 voices: Default, Power, Rock, Soft, Whisper.
  3. The Deluxe Edition contains nine voices: Nasal, Classic, Who, Special, plus the five voices of the Standard Edition.
  4. The standard version is priced at NT$2500 (US$79) and the deluxe version is priced at NT$3150 (US$99).
  5. The official launch will focus on selling the download version.
「いいね!」 4

His Nasal parameter sounds actually nasal? Nice! Gotta purchase him for synthesis of French.

「いいね!」 1

From Twitter page: The physical editions of the Oscar SV AI VoiceBank have been fully shipped overseas.
It will be available for download after December 28th!
The crowdfunded download version will be sent out on December 28th.
Oscar SV AI VoiceBank will be officially released on January 10th.

「いいね!」 3

OSCAR was released early on December 26! You can also get him as a bundle with Xia Yuyao~

https://twitter.com/VM_OscarAI/status/1739538488805212185

「いいね!」 3