SynthV Wiki
Advertisement
This project will not be released.

According to official sources this project has no current plans of release.
Please read the page for more information.

KAFU
SynthV Studio
CeVIO AI
|
Originals
|
Gallery

KAFU (可不) was a Japanese vocal that was under development by AH-Software Co. Ltd. and to be distributed by KAMITSUBAKI STUDIO;[1] she was expected to release as an AI voice database for Synthesizer V Studio in January 2024[2] but was delayed indefinitely[3] before being cancelled in January 2025.[4] She was originally released as a Song Voice for the CeVIO AI engine in July 2021. She is part of the Musical Isotope Project.

She is voiced by KAF, a Japanese virtual singer (VTuber) and performer.

Concept[]

KAFU is advertised as a "Musical Isotope" (音楽的同位体) of KAF, likely done in a way to separate the voice synthesizer character from the virtual singer.[5][6]

Her signature color is blue, specifically hex code #0000FF.[7]

Etymology[]

KAFU's name is a combination of "KAF", her voice provider, and "You", stylized as "U".

The kanji of her name, "可不" (KAFU) derives from "kanousei" (可能性) meaning "possibility," and "fukakai" (不可解) meaning "mysterious, incomprehensible." It was taken from the first character of each word respectively ("可"能性 and "不"可解; "ka"nousei and "fu"kakai.) In addition, "Fukakai," or "Incomprehensible" is a song sung by KAF, and it is one of her more notable songs.

Additionally, the name "KAFU" shares its romanization with KAF's character "Morisaki Kafu" (森先化歩) from KAMITSUBAKI CITY UNDER CONSTRUCTION. (神椿市建設中。) So far, this trait is also shared by the other characters in the Musical Isotope Project.

KAFU's full title is also known as "Musical Isotope KAFU" (音楽的同位体 可不).

Appearance[]

KAFU's appearance and theme are based on her voice provider, KAF, with many similar design elements and an overall white color-scheme, which was a recurring required design choice within the Musical Isotope Project.

KAFU's Synthesizer V version reused the same design as her previous CeVIO AI version. She wears a heavy white jacket with blue and black detailing and a hood similar to that of KAF's own attire. She also wears an over-the-knee length black dress, white tights, yellow socks, and blue-heeled short boots. Her hair is snow white and bobbed with low pigtails on either side, reminiscent of KAF's hairstyle, and is adorned with large diamond-shaped hair clips. Her eyes are dark blue and purple.

Relations[]

  • KAF - voice provider and basis for KAFU's character.
  • SEKAI - Fellow "Musical Isotope Project" vocal.
  • RIME - Fellow "Musical Isotope Project" vocal.
  • COKO - Fellow "Musical Isotope Project" vocal.
  • HARU - Fellow "Musical Isotope Project" vocal.

History[]

In October 2020, KAFU, a voice synthesizer character based on and voiced by virtual singer (VTuber), KAF, and illustrated by PALOW, was announced to be in production for CeVIO AI. KAFU's final voice was determined by KAF while taking account of the results from a questionnaire (showcasing voices "TYPE // A", "TYPE // B", and "TYPE // C" respectively), with KAF noting that while she was overall thrilled of the vocal as a VOCALOID fan herself and she had no concerns of KAFU "replacing" her, she felt as though some people looked at the two as inseparable entities that were indistinct from each other, prompting her to choose "TYPE // B" as the vocal to proceed with development.[8] After some delays due to the need to satisfy both KAF and fans, CeVIO AI KAFU released on July 7, 2021.[9]

2023[]

On July 7, 2023, at the end of the "V.I.P -Virtual Isotope Phenomenon- 1st" livestream, it was announced that KAFU would be in development for Synthesizer V AI and was expected to release in winter 2023.[10][11] On July 10, AH-Software Co. Ltd. confirmed that they would be developing KAFU though would not be publishing her themselves as this would be fulfilled by KAMITSUBAKI STUDIO.[1] On September 3, pre-orders were confirmed to begin the next day at 18:00 JST.[12]

On September 4, pre-orders for KAFU's physical packages were opened and would run until November 27 at 12:00 JST or if the available stock had sold out as she is a made-to-order product. She would be made available again depending on the number of products that could be produced, which KAMITSUBAKI STUDIO would announce on their official LINE account. KAFU's voice database could be purchased on its own or as a starter pack paired with the Synthesizer V Studio Pro editor. Her release date was confirmed to be January 15, 2024.[2]

On December 23, KAMITSUBAKI announced that her release date was being postponed due to various unspecified circumstances, with new release date being yet undetermined but to be announced as soon as details are decided. Pre-orderers were given the option to cancel their reservations due to the postponement, being instructed to contact the stores where they placed their preorders.[3]

2024[]

On May 14, 2024, it was revealed by KAMITSUBAKI that KAFU was indefinitely delayed due to her voice provider KAF being extremely uncomfortable with the realistic nature of Synthesizer V and with how it allowed KAFU to perfectly mimic her vocal traits on her published demos.[13]

2025[]

On January 19, 2025, PIEDPIPER from KAMITSUBAKI issued a statement definitely cancelling development of Synthesizer V KAFU. Apologies were given to everyone who was expecting her postponed release, and thanks were extended to Dreamtonics Co., Ltd. and AH-Software Co. Ltd. for their involvement in the development.[4]

Voice Database Information[]

Demonstrations[]

Demonstrations

フォニイ (phony) YouTube

Voice Databases[]

Product Information
Versions
  Lite Version/FLT Available?: No  Starter Pack Available?: Cancelled (Pre-order)  Web Version Available?: No

Package details as noted:
KAFU was a feminine Japanese voice database based on the singing style of KAF. As expected of an AI voice database, KAFU was capable of singing with a more human-like and realistic voice.

Vocal traits as noted:
With Diffusion Probabilistic Models (DPM), KAFU could sing more realistically through changes in volume and breath, further enriching singing details, making each singing unique, and providing creators with more options. Through this mode, she could use AI Retakes to create multiple "takes" of a sung section to find an ideal vocal without the need for "laborious tuning". This feature is only available in the Synthesizer V Studio Pro Editor.

Phonetic notes as noted:
  • Phoneme format: Romaji
  • Through Cross-lingual Singing Synthesis, KAFU was expected to sing in not only Japanese but also in English, Mandarin Chinese, Cantonese Chinese, and Spanish at the time of production, in addition to any other supported languages that would have been introduced afterwards had she been released. This feature is only available in the Synthesizer V Studio Pro Editor.

Vocal Modes as noted (Synthesizer V Studio Pro Editor only):
  • KAFU came with four variations:
    • Emotional
    • Kawaii
    • Stylish
    • Powerful

License agreements as noted:

Voice Libraries[]

Additional Information[]

Marketing[]

Throughout her time as a CeVIO AI vocal, KAFU had received official albums and merchandise. She was featured in live performances, including duets with her voice provider KAF, and was shown at the KAMITSUBAKI EXHIBITION, where various works like music videos and illustrations were on display.[14] Since November 17, 2021, works featuring KAFU were accepted on Piapro, a well-known Japanese center of the VOCALOID, and extensively, voice synthesizer community.[15]

Reputation[]

KAFU's character status
Reaction
As a CeVIO AI vocalist, KAFU became overwhelmingly popular and may be considered as one of the most notable CeVIO vocalists to date, with usages of over 7,000 songs at the time of the announcement of her Synthesizer V AI in July 2023.[16] Regarding her Synthesizer V AI announcement, she was met with surprise and excitement among fans.

Miscellaneous
KAFU's development for Synthesizer V AI was announced on July 7, 2023, coinciding with her second anniversary since her CeVIO AI release on July 7, 2021.

Achievements
  • First Synthesizer V product from THINKR INC.
  • First Synthesizer V from KAMITSUBAKI STUDIO.
  • First female Synthesizer V from KAMITSUBAKI STUDIO.
  • First Synthesizer V AI product to be cancelled.

References[]

Navigation[]

Advertisement