Synthesizer V AI ROSA (ロサ): Upcoming Japanese Female Voice from ZAN-SHIN Co., Ltd.

Can’t find a thread about the upcoming voice database for ROSA, a character from ZAN-SHIN Co., Ltd., so I’m making one to compile what we know so far. Feel free to use this thread for discussions~

ROSA’s been introduced since March 2021, when HirotoP confirmed that she was VOCALOID CUL’s younger sister (15 years old) and was themed pink. He also mentioned that ZAN-SHIN would like to produce an AI singing voice (later confirmed Synthesizer V AI) and a talk voice (later confirmed CeVIO AI), with the voice artist planned to be chosen via contest, which was planned for April. Regarding the vocal direction, he mentioned that it would be similar to how it was done for CUL with her VA Kitamura Eri. Also, due to interest from western fans, there were considerations to create English voices for both speech and singing (though given the circumstance of Cross-lingual Singing Synthesis, we don’t know if they’ll provide English samples or not).

After the contest concluded, Shirayuki was announced to be the chosen voice actress and recording sessions began in May. Throughout March and April, we’ve seen drafts of ROSA, but her final illustrations weren’t revealed until mid-May, with Yoshida Iyo being confirmed as the illustrator.

^First image is Synthesizer V design, second is CeVIO AI.

Later on in June, ROSA’s crowdfunding campaign for her CeVIO AI Japanese Talk Voice launched. This ran from June 3 to June 30. It was also announced that the plan was to release the voices in this order: Japanese Talk, Japanese Singing, English Talk, and English Singing (again, status is now unknown). The crowdfund concluded with 112% funded, prompting the production of her Japanese CeVIO AI voice.

In November, there was hope that production for Synthesizer V AI would begin in 2022. The CeVIO AI voice ended up releasing on December 17, with distributions handled by Internet Co., Ltd. (the company behind VOCALOIDs such as Gackpoid and Megpoid, and also helped release CUL)

We haven’t seemed to have heard much about ROSA’s Synthesizer V AI production for a while and a reference didn’t occur again until June this year. ROSA and WhiteCUL (a VOICEVOX project from ZAN-SHIN) ended up appearing in an online concert ( 柚子花主催LIVE -Planet Station- STAGE.3). There, their voice providers sang as the characters. HirotoP mentioned that hopefully the performances would give an idea of what an AI singing voice would sound like (please note that it wasn’t confirmed if WhiteCUL would be coming to Synth V. For now, assume this is about ROSA only).

It seemed that producing WhiteCUL for VOICEVOX had taken up ZAN-SHIN’s attention as she was in production for most of this year, starting from March. Her release is now approaching for September 30, and someone wanted to ask about ROSA’s status. They were told that ROSA’s Synth V project hasn’t started yet, but the plans are still active.

And that’s it for now~ What have you thought about ROSA so far? If you’ve heard her VA and her CeVIO AI voice, is she someone you are looking forward to? Feel free to discuss if you want to!


On May 31, Shirayuki tweeted that she’s been listening to and recording songs for ROSA (3 so far it seems). She said that this reminded her of the nightmare of recording for the CeVIO AI Talk software.

Don’t know if this is related to Synthesizer V AI or if these songs are simply going to be used for ROSA’s appearance at VVV MUSIC LIVE on June 25, but figured to bring it up anyways in case?


It’s been quite a long time since the original announcement, but ROSA’s crowdfund will finally begin mid-November.

The original talk voice crowdfund only barely reached its funding goal, but HirotoP is hopeful that the increased popularity of her character and voice provider in the time since will result in greater success for ROSA’s singing voice for SynthV Studio.

The last time We did a crowdfunding for ROSA , We had just debuted as a character, and ROSA was not well known at all, and even though she is CUL’s sister, she was also not well known in the talk community.

We managed to succeed in CF and made CeVIO AI TALK VOICE, but it has been less than 2 years since then. Thanks to the recent popularity of Ms. Snow, I think the recognition of ROSA has been gradually increasing.

And We think there are people who did not know about the previous CF and did not participate in the CF but are now supporting ROSA, and I am thinking of a way to give something back to them.

Those who support the current ROSA only know the talk ROSA! We have to make sure that the contents of CF will please those people as well!

Based on the reflection from last time, we are planning various courses for the start of CF in mid-November, so please support us!