Kizuna Akari (VOCALOID4)

Template:Infobox Vocaloid

This is an article about the Kizuna Akari software for the VOCALOID4 engine.

History

While celebrating the 5th anniversary of Yuzuki Yukari it was mentioned that the notion of making the next vocal was brought up. Having released all the planned vocals for Yukari both in Vocaloid and Voiceroid, the issue became a more serious one. Bumpyうるし mentioned in an interview that he had been wanting to do the next vocal for some time, but focus was always on expanding the Yukari vocal before then.

Akari was a secret project and nothing was ever shown to the public. Each member of the team worked independently, bringing in voice actors and singers to interview, eventually resulting in 10 candidates. They noted, as the character and voice would grow, this would be a long term relationship between the team and the candidate. Ah Software provided advice, songs were recorded in the voice and tested to see how they sounded. The audition took 3 days to carry out, resulting in a 4 day session. Though they knew what they were after no on-the-spot decisions could be made.

Temporary voicebanks were made, these have been known to occur in the past with other Vocaloid developments and act as tests on how a voice fairs within the Vocaloid software. Both Vocaloid and Voiceroid were sampled at the same time to make a overall judgement of the voice. Since Ah Software had 7 years of experience of Vocaloid and Voiceroid by this point, the process was fairly efficient overall.

Madoka Yonezawa, who was chosen to be the provider, had a music background favouring singing and had studied voice acting. Both talents of which she is able to use together. Yonezawa had heard of Hatsune Miku due to the subject being discussed as it was considered a sensitive matter among her peers and had an interest in Vocaloid. She became acquainted with Kagamine Rin as she went karaoke with Kumiko Ito who often sung Rin's songs and eventually she sung a Len song live. More and more Vocaloid songs appeared, and since DTM noted anyone could write a hit song with Vocaloid, a lot of new unknown talent was appearing. However, voice acting at the time was a largely narrow field and she was confused by some of the information that existed in regards to Vocaloid. It was noted it wasn't until Gackt leant his voice to Camui Gackpo that she suddenly considered one day making a Vocaloid herself as Gackt had noted how Vocaloid was no longer just Hatsune Miku and Kagamine Rin & Len.

Madoka Yonezawa did not fear her existence of a singing and talking voice would deprive her of work as she was seeing a lack of work at the time. When she voiced Akari she imagined the character of "Kizuna Akari" as both a singer and a speaker that could lead to work with anime and video games. Yonezawa saw the chance to voice Akari as an expansion of work, rather then a deprivation of work for this reason. At the time she had contact with the head of Vocalomakets. During her audition she sung only her favourite songs and practiced them. Some of the script was hard to do and she did not know what certain things were such as what the pitch "Tonkatu" (トーンカツー) meant. She did the performance with a sort of cutesy vocal and did not know at the time the recordings were being used to test a Vocaloid, extracting consonants and vowels for creation of a basic voicebank. The script wasn't what she expected as she thought she would be recording such as "A" (あ), "I" (い), "U" (う), "Ka" (か), "Ki" (き), "Ku" (く).Yonezawa's audition was in November 2016. She was contacted 3 months later and was told it would be judged based on her performance based from "Tonkatu". She was one of 10 people whose samples had been made into a basic voicebank. When she was told she had made it was overjoyed at being chosen and the choice had been made itself fairly quickly.

Different patterns of recording were tried at the time to see what was best.The name of the Vocaloid "Kizuna Akari" had already been chosen by recording, but was not yet the finalised name and the type of character she would be would also undetermined as they had to see what work best for Yonezawa'a voice. She listened to the song "Red Dragonfly" (赤とんぼ) and picked up what Vocaloid was again. She was embarrassed by her early singing results, but as discussions occurred the voice that transformed into Akari was fluffy sounding. The voice needed to be narrowed down for this reason, in total 5 versions of the voice were made.

From the beginning it had to be made clear Akari was a different vocaloid to Yukari. When provisionary voicebanks was made, demo songs were also produced to help the evolution of the development of the Vocaloid along, though ideas had started to form such as titles. When everyone's opinions were gathered, a overall decision was made on things. As with Yukari before her, Akari's voice was made before they even considered the character itself.

The illustrator wasked if they could do the same as they had done with Yukari for Akari's artwork. Listening to the voice of the character, and seeing the characters side by side, the name was picked out about this time. Since Yukari was based on the moon, they kept the astrological thing discussing on if Akari should be based on the sun or stars. The name had to hold a story behind it. "紲" meaning "Bond" was problematic for them because it could be mistranslated to "tying people" and they also considered forcing the word "Star" into the name itself. Eventually they decided on on a name.

Development of the Voiceroid and Vocaloid took place during this, with the Voiceroid recording session taking 2 days and being done in a method that resembled the provisionary voicebank for Vocaloid. Vocaloid, though already one step ahead the Voiceroid, took a further 6 days to record, this is why the Voiceroid was released ahead of the Vocaloid voice. Another day was taken to record exVoice samples for both. It was originally intended for the two to be released on the same day, but the Vocaloid was taken far longer to produce, being 4 months behind her scheduled release of her Voiceroid. Speaking on the exVoice, the issue with Vocaloid is that consonants feel weak at times or cannot produce alternative consonants. The exVoice for her Vocaloid release is intended therefore to be able to be copied and pasted on the sound track in place of Vocaloid alternatives.

On November 30, 2017, Vocalomakets and AH-Software Co. Ltd. revealed a new character to be added to the VOICEROID2 line: Kizuna Akari. Her VOICEROID2 product was set to be released on December 22 that year. It was also mentioned that they were working diligently on her VOCALOID version at the time. Juu Ayakura was confirmed to be the illustrator for Akari, but her voice provider was not to be revealed until the VOCALOID package was released. Vocalomakets later tweeted that there is no information about Akari's relation to her predecessor, Yuzuki Yukari, and only confirmed parts of her biography.

On December 17, Vocalomakets hosted a livestream featuring new Yuzuki Yukari merchandise and three Akari VOCALOID demos. It was revealed that Akari also has exVOICE files and that she was expected to release in 2018.

On March 29, 2018, Akari's preorder period was launched. It was revealed that Akari would be available as an individual pack (physical and digital copies), a starter pack, and a dual pack with Yukari. Her release date was confirmed to be April 26, 2018.

On April 13, Akari's trial was made available.

On June 3, Akari's voice provider was confirmed to be Madoka Yonezawa.

Product Information

Demonstrations

Demonstrations

Sprites!☆	NicoNico YouTube SoundCloud
リナリア (Linaria)	NicoNico YouTube
あなたと二人で (Anata to Futari De)	NicoNico YouTube
exVOICE Examples	NicoNico YouTube
パニジャパーの奇跡 (Panijapaa no Kiseki)	NicoNico YouTube SoundCloud
星のオーケストラ (Hoshi no Orchestra)	SoundCloud
オリオンに願いを (Orion ni Negai o)	SoundCloud

Recommended

• OS:
  • Windows: Windows 8.1, Windows 8, Windows 7 (32/64bit)
  • Mac: OS X 10.8 (or later)
• CPU: Intel Dual Core CPU
• RAM: 2GB or more
• Hard Disk Capacity:
  • Standalone package:
    • More than 1.5GB (using VOCALOID4 Editor)
    • More than 9.5GB (using VOCALOID4 Editor for Cubase + Cubase)
  • Starter Pack: More than 1.65GB (using VOCALOID4 Editor)
  • Yukari and Akari dual package
    • More than 2.5GB (using VOCALOID4 Editor)
    • More than 10.4GB (using the VOCALOID4 Editor for Cubase + Cubase)
• Internet connection.

Voicebank Libraries

Product Information
Optimum
Optimum Range: G2 ~ A4  Optimum Tempo: 70 ~ 190 BPM   Total Tempo (min-max): 120 BPM  No. of Keys: W ~ 16, B ~ 11, Total ~ 27
Versions
Trial/Demo Vers?: Yes  Starter Available?: Yes
Package details as noted:
Akari is meant to sound kind and gentle with the intention to portray a cute, bright female voice and under the theme of "starlight". Her package also comes with an "exVoice" set. There is a voice release for VOICEROID2.
Vocal traits as noted:
She is able to sing a side range of genres from rock to pop. While Yuzuki Yukari is meant to be an adult voice, Akari is a more cutesy vocal with a lighter tone. At 70bpm, Akari can handle slow songs relatively well, but not as well as Yukari at 60bpm. There are a few genres of music that Yukari can handle better than Akari. Her tempo range is superior overall compared to Yukari and is double the size range. As a result, she can handle far more genres of music. Aside from the slower song genres Yukari can handle, Akari covers all of Yukari's tempo range and has a superior range in total. Akari sounds best when going from low to high. The team behind her believe she is good for beginners.
Voicebank sample
Kizuna Akari VOCALOID4
Cross-Synthesis as noted:
Kizuna Akari XSY group This vocal will register for XSY, however this is a single release voicebank and there are no additional vocals included for her to utilize this function. Therefore users will have to purchase additional voicebanks not belonging to Akari for use of XSY from her XSY group "AHS". From Ver.4.3.0 of the VOCALOID engine onwards, a XSY group "AHS" was added to VOCALOID. All vocals within the "AHS" group can XSY with each other. This vocal release is part of this group. If a user owns one or more vocal releases within the "AHS" group, XSY between them will open up. The group has 8 characters within it that can XSY: Yuzuki Yukari, Macne Nana, Macne Petit, Tohoku Zunko, Hiyama Kiyoteru, Nekomura Iroha, SF-A2 miki and Kizuna Akari. The "AHS" group is currently the second largest XSY group within VOCALOID, having a total of 14 voicebanks that can XSY with each other. If old versions that have since been re-released are removed, there are currently 12 unique voicebanks. These are: "Yuzuki Yukari Jun", "Yuzuki Yukari Onn", "Yuzuki Yukari Lin", "SF-A2 miki", "Tohoku Zunko", "Macne Nana", "Macne Petit", "Kizuna Akari", "Hiyama Kiyoteru Natural", "Hiyama Kiyoteru Rock", "Nekomura Iroha Natural" and "Nekomura Iroha Soft". In total if a user was to own every VOCALOID3 and VOCALOID4 AH-Software vocal within the "AHS" group, this offers the equivalent of 182 additional voicebanks achievable via XSY. The total tone variation offered by the combined packages comes to a theoretical 196 voicebanks in total. Note that also "GWL" does not become viable for use with any V3 vocals used as the primarily vocal. This should be considered when using any vocal for XSY within the "AHS" group from V3. Also note that the updated vocals that brought a VOCALOID from VOCALOID3 into VOCALOID4 provided few differences, aside from adding GWL. Therefore these updates can provide little difference when used for XSY.
view • edit

Navigation

Template:Nav ahsoft