Jump to content

Draft:Synthesizer V

From Wikipedia, the free encyclopedia
  • Comment: Twitter is not a reliable source. Neither are pixiv or weibo for that matter. These websites consist of user-generated content and cannot be considered independent, secondary source for determining its notability. I would recommend these references to Twitter/pixiv/weibo be removed, and replaced with reliable sources that provide significant coverage of this topic if it is reasonable to do so. Utopes (talk / cont) 20:22, 13 September 2023 (UTC)


Synthesizer V Studio (also known as Synthesizer V, or Synth V) is a singing voice synthesizer (SVS) developed and distributed by Dreamtonics Co., Ltd. which was first released in December 2018.[1] The software is designed for the creation of artificial singing vocals for music production but is also widely used for artificial speech. Officially supported languages include English, Japanese, Mandarin Chinese, Cantonese Chinese, and Spanish.[2] It became popular among SVS enthusiasts after the release of Synthesizer V AI, which is notable for its use of neural networks, which aid the software in producing more natural sounding results than the traditional concatenative methods used in most SVS programs.

Synthesizer V Editor (Generation 1)

[edit]

[3] Currently known as Generation 1, R1 or gen1, Synthesizer V Editor was released by Kanru Hua in August 2018, selling itself as a simple-to-use modern vocal synthesizer.

History

[edit]

Although Synthesizer V's first line of code has been dated back to 2015, as part of the proto-proto libllsm, which would become part of SynthV and the UTAU resampler Moresampler[4][5], the first signs of official development began in 2017, with the release of a track titled "Synthesizer V Demo" by Kanru Hua on SoundCloud.[6] It features three voices: ENG-F1 (later released as Eleanor Forte)[7] singing "Young and Beautiful," JA-F1 singing "Yuki no Hana," and MAN-F1 singing "Mo Li Hua." The project was unveiled to the public on the 1st of December 2017 and was well received.[8]

August 19th, 2018, a technical preview version of the software was released for user feedback.[9][10] December 24th, the official production version was released, alongside AiKO and Genbu. The software received updates until its retirement in July 2020, due to the release of Synthesizer V Studio.

Available Voices[1]

[edit]
Voices produced by VOLOR[11]
[edit]
Name Release Date Language Voice Provider Avatar Illustrator Overview
Eleanor Forte[12] 19/07/2018 English Unknown Sleepy Ho11ow[13] Synthesizer V's first voice. Its Standard version was released alongside the technical preview of the software. Its character is described as honest, endeavoring, with an unwavering sense of morality.
AiKO[14] 28/12/2018 Mandarin Chinese Liang Xiao Miao[15] LAM Synthesizer V's first Chinese voice. The character was originally created as Animen's (Parent company of VOLOR) mascot in 2012 and Liang Xiao Miao was the character's voice actress much before SynthV's release.[15] Its character is described as enthusiastic about everything and careless.
Genbu[16] 24/12/2018[17] Japanese Garo[18] onidaruman[19] Synthesizer V's first male Japanese voice. Its character is described as unpolished on the outside, but calm and caring.
Yamine Renri[20] 11/11/2018 Japanese Yuzuri[21] kgr[22] Synthesizer V's first female Japanese voice. Yamine Renri was originally created as a voicebank for UTAU, but received a Synthesizer V version.

Voices produced by Quadimension[23]

[edit]
Name Release Date Language Voice Provider Avatar Illustrator Overview
Chiyu[24] 10/04/2019 Mandarin Chinese YouMao UNEKO[25] ATDAN[26] All Quadimension voices are based on characters for the company's fantasy project, MEDIUM⁵. Chiyu's character is described as brave and honest.
CangQiong[27] 09/07/2019 Mandarin Chinese Kinoko Mogu[28] ATDAN All Quadimension voices are based on characters for the company's fantasy project, MEDIUM⁵. CangQiong's character is descirbed as logical and kind.
Shian[29] 20/06/2019 Mandarin Chinese warma[30] ATDAN All Quadimension voices are based on characters for the company's fantasy project, MEDIUM⁵. Shian's character is described as childish, soft and warm.
Haiyi[31] 06/09/2019 Mandarin Chinese Huazheng Anly[32] ATDAN All Quadimension voices are based on characters for the company's fantasy project, MEDIUM⁵. Haiyi's character is described as chatty and mischievous.

Updates

[edit]
Update Release date
Synthesizer V Editor update (Build 013) 27/12/2018
Synthesizer V Editor update (Build 014) 04/01/2019
Synthesizer V Editor update (Build 015) 10/01/2019
Synthesizer V Editor update (Build 016) 12/03/2019
Synthesizer V Editor update (Build 017) 02/05/2019
Synthesizer V Editor update (Build 018) 01/09/2019

Web Synthesizer V[33]

[edit]

Web Synthesizer V is a browser-run stripped down version of Synthesizer V Studio, initially released in the 31 of December 2019[34]. It exists as an avenue for users to test new updates before they are applied to the main engine[33]. It only ever received one update, on July 22, 2020 the software received a bigger selection of voices and support for different browsers.[35]

Available Voices

[edit]
Name Release Date Language Avatar Illustrator Voice Provider Overview
Saki 31/12/2019 Japanese n/a n/a Initially referred to as JA-F2, Saki was the first vocal available for both Web Synth V and Synth V Studio. The name was later changed to Saki for the release of generation 2.
Kotonoha Akane & Aoi 22/07/2020[35] Japanese Yui Sakakibara[36] Yoshida Yoshitsugi[37] Despite being a singular voice, Kotonoha Akane & Aoi is representative of two distinctive characters. It was made available for Web Synthesizer V through a collaboration with AH-Software Co. Ltd., the company responsible for the voice.[35]

Synthesizer V Studio (Generation 2)

[edit]

Synthesizer V Studio, also known as generation 2, R2 or gen2, is the current release of Synthesizer V. It is available in 2 versions, "Synthesizer V Studio Pro" initially released July 10, 2020 and "Synthesizer V Studio Basic" initially released July 30, 2020. It features a completely reworked engine, which combines sample-based synthesis and AI machine learning. It was through this version that Synthesizer V AI (also known as Generation 3) was released, an update which allowed certain voices to take full advantage of the software's AI technology, which has since become Synth V's most popular feature.

Synthesizer V Basic and Synthesizer V Studio differ in that Basic is a stripped down version of the engine available for free. Basic lacks access to some of Pro's features, only allowing the creation of 3 tracks per project, and prohibits commercial use, requiring a user to upgrade to Pro should they wish to sell any work made with the engine. Voices sold for Synthesizer V have a similar model, providing "Lite" versions of voice libraries that are available for free, but are lower quality, have fewer features, and do not permit users to sell work made with them.

History

[edit]

The software was initially unveiled by AH-Software Co. Ltd. on their official Twitter account June 26, 2020, with the voices of Saki and Kotonoha Akane & Aoi.[38] The software was released next month, on July 10, 2020 to Chinese users through Beijing Photek S&T Development Co., Ltd.'s Taobao and on July 15, for international users through AniCUTE. Later on July 30, the Basic version was released for Japanese users through AH-Software Co. Ltd, and the next day, it was made available on Dreamtonics' official website.

On December 25th, 2020 AI support was implemented into Synthesizer V Studio through Synthesizer V AI, alongside the release of Saki AI. From this point voices would be categorized as Standard Voices, and AI Voices, according to whether or not they use Synth V AI.

On December 23rd, 2021, cross-lingual synthesis was implemented into Synthesizer V Studio for all AI voices, effectively rendering all AI voices trilingual regardless of their native language, allowing them to sing in Japanese, Chinese or English. This feature is exclusive to the paid versions of AI voices running on Synthesizer V Pro. Some voices, like Stardust Infinity, have additional recordings in more than one language, as this allows for more fluent results when using cross-lingual synthesis.

On February 2, 2022, AH-Software Co. Ltd. announced that Synthesizer V Studio sales had been exceeding the their expectations, marking the software's growth in popularity.[39]

On February 27th, 2023, Dreamtonics announced plans to support Cantonese cross-lingual synthesis, a dedicated synthesis mode that allows vocalists to rap, and two new vocalists dedicated to rapping. As of June 2023, all features related to rap are only available when a vocalist is using English or Mandarin Chinese, however Dreamtonics has announced that they intend to add Japanese compatibility to the feature in a future update.[40]

On November 5th, 2023, third-party developer Eclipsed Sounds announced that Synthesizer V would support Spanish cross-lingual synthesis with the release of version 1.11.0 of Synthesizer V AI. This was the first time that support for a language had been implemented by a third-party developer in cooperation with Dreamtonics. Th

Available Voices[41]

[edit]

Although voices made for Synthesizer V Editor were also made available for Synthesizer V Studio, many others were produced with the new technology in mind.

Voices produced by Dreamtonics

[edit]
Name Release Date Native Language Gender Voice Provider Overview
Saki December 31, 2019 Japanese Female Unknown The first voice produced by Dreamtonics, Saki is a female voicebank available both as standard and AI. It was made to be natural and mature.
Qing Su December 1, 2021 Mandarin Chinese Female First Mandarin Chinese vocal made available for Synthesizer V Studio. Qing Su is intended to be versatile within the realm of C-Pop, and has a mature, natural tone.
Kevin February 4, 2022[42] English Male One of two first male voices released by Dreamtonics for Synthesizer V Studio. Made to be clear powerful, with a wide range.
Ryo Japanese Male Released alongside Kevin, Ryo serves as his counterpart, made to be more gentle but still flexible.
Mo Chen April 9, 2022[43] Mandarin Chinese Male Synthsizer V's first male voice made to sing in mandarin Chinese. Made with C-Pop in mind, he is a softer vocalist.
An Xiao May 4, 2022[44] Mandarin Chinese Male Made with expressiveness in mind, An Xiao has a powerful, deeper vocal.
Feng Yi May 29, 2022[45] Mandarin Chinese Female Made with expressiveness in mind, Feng Yi has a clear, bright tone. She is notable for her capability to sing in the style of Peking Opera through her "Opera" vocal mode, but she is not completely dedicated to the genre.
Natalie July 10, 2022[46] English Female Intended to have a soft tone with wide range, Natalie is Dreamtonics' first English female voice.
Yuma Japanese Male Released alongside Natalie, Yuma is a male vocal with a youthful, androgynous tone made with expressiveness and power in mind.
Weina July 21, 2022[47] Mandarin Chinese / English Female Weina Hu First voicebank to be released for the Gold Series, with a higher production standard compared to Dreamtonics' other voices. Weina is notable for having complete recordings for both Mandarin Chinese and English, however her English capabilities are restricted to Synthesizer V Pro users, despite not making use of cross-language-synthesis.
Mai October 11, 2022[48] Japanese Female Kitta Honoka Made with J-Pop in mind, Mai has a crisp and bright tone. She was the first Dreamtonics voice to be made fully available for free to users who bought the Pro version of the software. She
Xuan Yu December 12, 2022 Mandarin Chinese Male Unknown Made with rock music in mind, Xuan Yu has a powerful, warm tone.
Cong Zheng Mandarin Chinese Female Made with rock music in mind, Cong Zheng has a solid, mature tone.
Cheng Xiao January 9, 2023 Mandarin Chinese Female Cheng Xiao is a vocalist made with Peking Opera in mind. Unlike Feng Yi, who was also capable of singing Peking Opera, Cheng Xiao is fully dedicated to the genre.
Ninezero February 16, 2023 English Male NINEZERO Ninezero is a vocalist meant for rock and metal music. He has a powerful and expressive tone. He also features a small library (referred to as a pitagoe) that contains about 200 short recordings of his voice provider saying common phrases.
D-Lin July 4, 2023 Mandarin Chinese Male Unknown D-Lin was one of two vocalists unveiled on February 27th, 2023. He is a vocalist dedicated to rapping. A free trial of his voice is available to Synthesizer V Pro users who participate in public betas.
Ritchy July 4, 2023 English Male MBanja Ritchy Ritchy, like D-Lin, was unveiled on February 27th, 2023, and is also a vocalist dedicated to rapping. A free trial of his voice is available to Synthesizer V Pro users who participate in public betas.

Voices produced by Eclipsed Sounds

[edit]
Name Release Date Language Gender Voice Provider Overview
Solaria January 19, 2022 English Female Emma Rowley Solaria (Originally named Solaris) is Eclipsed Sounds' first released voice, made with expressiveness and realism in mind. Its production was crowdfunded on Indiegogo.
Asterian December 10, 2022 English Male Eric Hollaway Made as a counterpart to Solaria, Asterian was made to be equally expressive and realistic, but with a more unique bass tone.
Saros November 24, 2023 English Neuter Unannounced Made as a "midway point between SOLARIA and ASTERIAN", as well as a "very strong and expressive voice", SAROS is a tenor voice with a uniquely strong high range compared to existing tenor vocals. They are also the first voice on Synthesizer V to have a non-binary gender identity, referred to in promotional materials solely with they/them pronouns.

Voices produced by AUDIOLOGIE

[edit]
Name Release Date Native Language Gender Voice Provider Overview
ANRI December 11, 2021 English Female kurapichu
JUN April 20, 2023 English Male Lavy
ANRI Arcane December 11, 2023 English Female kurapichu ANRI Arcane is an updated version of ANRI, featuring additional data for cross-lingual synthesis, higher quality recordings, and distinct vocal modes.

Voices produced by Beijing Photek

[edit]
Name Release Date Native Language Gender Voice Provider Overview
Xingchen Infinity February 20, 2022 Mandarin Chinese Female Chalili
Yongye Minus December 8, 2023 Mandarin Chinese Female Sawako Suihuazi

References

[edit]
  1. ^ a b "Synthesizer V | Dreamtonics株式会社". Dreamtonics (in Japanese). 2020-04-15. Retrieved 2022-11-03.
  2. ^ "Spanish is Coming to Synthesizer V Studio!". 6 November 2023.
  3. ^ "Synthesizer V - Official Website". Synthesizer V - Official Website. Retrieved 2022-11-03.
  4. ^ "https://twitter.com/khuasw/status/1112006856283570177". Twitter. Retrieved 2022-11-02. {{cite web}}: External link in |title= (help)
  5. ^ Kanru, Hua (2018-10-08). "Moresampler – Kanru Hua's Website". Archived from the original on 2018-10-08. Retrieved 2023-01-23.
  6. ^ Synthesizer V Demo, retrieved 2022-11-02
  7. ^ "https://twitter.com/khuasw/status/1076431855992225792". Twitter. Retrieved 2022-11-02. {{cite web}}: External link in |title= (help)
  8. ^ Unveiling Synthesizer V, the New Virtual Singer Software, 30 November 2017, retrieved 2022-11-02
  9. ^ "https://twitter.com/khuasw/status/1027354095227523072". Twitter. Retrieved 2022-11-02. {{cite web}}: External link in |title= (help)
  10. ^ "https://twitter.com/khuasw/status/1031446551732707328". Twitter. Retrieved 2022-11-02. {{cite web}}: External link in |title= (help)
  11. ^ "ANiCUTE". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  12. ^ "Eleanor Forte AI (English)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  13. ^ "EleanorForte, SynthesizerV, SynthV / ■Eleanor Forte/エレノア・フォルテ". pixiv. Retrieved 2022-11-03.
  14. ^ "AiKO Sound Library (Chinese)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  15. ^ a b "Animen". company.animen.com.tw. Retrieved 2022-11-03.
  16. ^ "GENBU Sound Library (Japanese)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  17. ^ "VirVox Projectキャラクターのプロフィール｜VirVoxプロジェクト(旧ページ)｜note". note（ノート） (in Japanese). 18 June 2021. Retrieved 2022-11-03.
  18. ^ "https://twitter.com/aman0_kei/status/1283582563664056321". Twitter. Retrieved 2022-11-03. {{cite web}}: External link in |title= (help)
  19. ^ "SynthesizerV, SynthV / SynthV「ゲンブ」 / December 23rd, 2018". pixiv. Retrieved 2022-11-03.
  20. ^ "SynthV". 闇音レンリ official site (in Japanese). Retrieved 2022-11-03.
  21. ^ "PROFILE". 闇音レンリ official site (in Japanese). Retrieved 2022-11-03.
  22. ^ "UTAU, supporting UTAU, black tights / 【UTAU】闇音レンリ立ち絵イラスト". pixiv. Retrieved 2022-11-03.
  23. ^ "平行四界Quadimension - BOOTH". quadimension.booth.pm. Retrieved 2022-11-03.
  24. ^ "赤羽 Chiyu 歌声ライブラリー (中国語)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  25. ^ "Sina Visitor System". passport.weibo.com. Retrieved 2022-11-03.
  26. ^ "girl, Akabane, SynthV / Fire". pixiv. Retrieved 2022-11-03.
  27. ^ "蒼穹 Cangqiong 歌声ライブラリー (中国語)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  28. ^ "Sina Visitor System". passport.weibo.com. Retrieved 2022-11-03.
  29. ^ "詩岸 Shian 歌声ライブラリー (中国語)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  30. ^ "Sina Visitor System". passport.weibo.com. Retrieved 2022-11-03.
  31. ^ "海伊 Haiyi 歌声ライブラリー (中国語)". ANiCUTE (in Chinese (Taiwan)). Retrieved 2022-11-03.
  32. ^ "Sina Visitor System". passport.weibo.com. Retrieved 2022-11-03.
  33. ^ a b "Web Synthesizer V". Synthesizer V - Official Website. Retrieved 2022-11-11.
  34. ^ "https://twitter.com/khuasw/status/1212025665823723522". Twitter. Retrieved 2022-11-11. {{cite web}}: External link in |title= (help)
  35. ^ a b c "https://twitter.com/dreamtonics_en/status/1285837842564825088". Twitter. Retrieved 2022-11-11. {{cite web}}: External link in |title= (help)
  36. ^ "Synthesizer V 琴葉 茜・葵｜製品情報｜AHS(AH-Software)". www.ah-soft.com. Retrieved 2022-11-11.
  37. ^ "SEKITOU | Yoshida Yoshitsugi". sekitou.sub.jp. Retrieved 2022-11-11.
  38. ^ "https://twitter.com/ahsoft/status/1276351253073629185". Twitter. Retrieved 2022-11-11. {{cite web}}: External link in |title= (help)
  39. ^ "待望の男性歌声データベース2種類がついに登場！『Synthesizer V AI Ryo』 『Synthesizer V AI Kevin』本日発売開始｜AHS(AH-Software)". www.ah-soft.com. Retrieved 2022-11-11.
  40. ^ "Synthesizer V Studio 1.9.0b1 Update: Rap, Cantonese and More | Dreamtonics株式会社". Dreamtonics (in Japanese). 2023-04-18. Retrieved 2023-06-01.
  41. ^ "Synthesizer V | Dreamtonics株式会社". Dreamtonics (in Japanese). 2020-04-15. Retrieved 2023-01-05.
  42. ^ "https://twitter.com/dreamtonics_en/status/1489434512337809408". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  43. ^ "https://twitter.com/dreamtonics_en/status/1512718709864820736". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  44. ^ "https://twitter.com/dreamtonics_en/status/1521791929092292608". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  45. ^ "https://twitter.com/dreamtonics_en/status/1530745841727598592". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  46. ^ "https://twitter.com/dreamtonics_en/status/1578264447852105728". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  47. ^ "https://twitter.com/dreamtonics_en/status/1550055032535777282". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
  48. ^ "https://twitter.com/dreamtonics_en/status/1590600959725830146". Twitter. Retrieved 2023-01-05. {{cite web}}: External link in |title= (help)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Draft:Synthesizer_V&oldid=1257940150"