(January 2004)
AI: Japanese, English, Chinese
(February 2023)
VOCALOID(Japanese:ボーカロイド/ボカロ, Chinese: Bógēlè, English abbreviation: VOCALO) is a collective term for the electronic singing synthesis technology developed by Yamaha Corporation of Japan, as well as the software that applies this technology.
Introduction
VOCALOID is an electronic singing synthesis software independently developed by Yamaha Corporation of Japan. By inputting melody and lyrics, it can synthesize singing voices close to human voices. As of December 2022, the program is divided into six generations: VOCALOID (V1), VOCALOID 2, VOCALOID 3, VOCALOID 4, VOCALOID 5, and VOCALOID 6 (VOCALOID: AI). The software's traditional concatenative voicebanks were initially based on the "Frequency-domain Singing Articulation Splicing and Shaping" technology developed by the Music Technology Group (MTG, an institution under Pompeu Fabra University) and licensed to Yamaha. It collects voice samples from different humans and then creates singing voice databases (voicebanks) to synthesize sound. VOCALOID: AI uses neural network synthesis technology to achieve more natural synthesized singing voices.
Origin of the Name
The name VOCALOID is formed by combining "vocal" (singing) with the suffix "-oid". The suffix "-oid" means "similar to" or "in the shape of". When combined with "human" (human), it becomes "humanoid" (having human characteristics, humanoid robot); when combined with the Greek root for man "andro-" (or "andr-" before a vowel), it becomes "android" (robot). Thus, VOCALOID means "mimic singing voice", "a created thing similar to a human vocal voice", or "something that can sing".[1]。
However, there is another explanation for the origin of the name: "Vocal·Android (Singing·Robot) = VOCALOID (ボーカロイド)", meaning "a robot that can sing".[2]。
Translated Names
Official translated name "Bógēlè", however, except for the Mainland China version of 《Taiko no Tatsujin》, almost no one uses it。
In China, it is commonly referred to by the public as "VFamily," which is also a homophone for "V+". If translated fully into Chinese based on its etymology, it could be called "Singing Robot."A literal translation would result in the strange "vocalic semi-consonant (like r)" (i.e., "vocaloid" in the field of phonetics. Correspondingly, "vocal" means "vocalic").
Additionally, because the Japanese abbreviation for Vocaloid, "ボカロ” resembles theKanji characters "术力口", Vocaloid is also commonly referred to as "Shu Li Kou" by fans. This term is also frequently used when referring to concepts related to Vocaloid, such as "Shu songs," "Shu circle," "Shu person," etc.
Release Date
| Version | Release Date | End of Support Date |
|---|---|---|
| VOCALOID | January 15, 2004 | January 1, 2014 |
| VOCALOID 2 | June 29, 2007 | March 31, 2016 |
| VOCALOID 3 | October 21, 2011 | Supported |
| VOCALOID 4 | December 17, 2014 | |
| VOCALOID 5 | July 12, 2018 | |
| VOCALOID 6 | October 13, 2022 | On Sale |
| VOCALOID β-STUDIO VX-β[3] | August 22, 2023 | April 1, 2024 |
Parameters
Simply inputting lyrics and melodies to synthesize singing will result in a mechanical sound. The VOCALOID editor provides users (commonly known as "tuners" or "Producers") with many quantifiable parameters to enrich the details and emotions of the singing voice.
Among the currently commonly used editors, V4 has 12 parameters; V5 removed the XSY parameter and added two new parameters, AIR and EXT, for a total of 13; V6 (version 6.5 and above) restored the XSY parameter, for a total of 14; VOCALOID:AI has only 3 parameters (increased to 4 with the addition of AIR in version 6.5 and above). The details are as follows:
* Note: The abbreviation is before the full English name of the parameter, and the full name is after. If there are two sets of English names (e.g., Opening/Mouth), the former is the parameter name for V4 and earlier versions, and the latter is for V5 and later versions. Some parameters have the same full English name but changed abbreviations in V6 (e.g., Pitch Bend, PIT/PB).
- Velocity (VEL, Velocity)
- Range: 0–127 Default: 64 (Non-continuous parameter)
- InfluenceConsonantlength. The higher the value, the shorter the consonant duration and the more hurried the pronunciation; the lower the value, the longer the consonant duration, making the pronunciation feel more powerful due to the pause.
- For non-plosives, it affects the overall length of the consonant; for plosives, it affects the length of the closure (silent gap) before the release.
- Dynamics (DYN, Dynamics)
- Range: 0–127 Default: 64 (Supported by AI voicebanks)
- Affects volume. The higher the value, the louder the volume. It can create crescendo or decrescendo effects, and can also be used for "consonant/vowel swapping" through multi-track coordination. It is recommended to use the Mixer (shortcut "F3") for overall track volume adjustment.
- The effect curve of DYN is a relatively smooth piecewise function. When DYN < 36[4], the change in loudness (dB) is logarithmic[5]; when DYN > 36, the change in loudness (dB) is linear[6]。
- Breathiness (BRE, Breathiness)
- Range: 0–127 Default: 0
- Controls the proportion of noise (breathiness) in the singing voice. The higher the value, the louder the breath sound. This parameter was reconstructed in V5 to make the breath effect more realistic.
- Brightness (BRI, Brightness)
- Range: 0–127 Default: 64
- InfluenceVowels (Finals)components of mid-to-high frequencies (mainly around 3k-6k), which also affects volume. The higher the value, the stronger the sense of brightness; the lower the value, the more steady and gentle it becomes, which can enrich the emotion of the singing voice. Values that are too low may cause the sound to become muffled.
- The response curve of BRI is a piecewise function with distinct inflection points[7]. When BRI < 64, the change is more pronounced[8], when BRI > 64, the change is less pronounced[9]。
- Clearness (CLE / CLR, Clearness)
- Range: 0–127 Default: 0
- Adds gain to the ultra-high frequency band (approximately 8k and above). The higher the value, the stronger the sense of clarity; the lower the value, the deeper the sound. Using values that are too high may lead to excessive sibilance.
- Opening (OPE, Opening / MOU, Mouth)
- Range: 0~127 Default: 127 (Non-continuous parameter)
- Simulates the degree of mouth opening and closing, adjusting pronunciations where the mouth shape is too wide. The default parameter is the maximum value. The lower the parameter, the smaller the mouth opening when the singer performs.
- Gender Factor (GEN, Gender Factor / CHA, Character)
- GEN Range: 0~127 Default: 64 CHA Range: -64~+63 Default: 0
- Affects the formants of vowels to change the timbre. In V4 and earlier versions, the higher the value, the stronger the masculine feel; the lower the value, the more it becomes a female or even a child's voice. In V5 and later versions, it is the opposite. If the value is significantly increased or decreased, it can sound completely different from the original voice.Extremes in either direction are quite terrifying; use with caution.
- When the gender parameter leans towards masculinization, the pronunciation tends tosmaller mouth opening, heavier nasal sound, rounder mouth shape, and back articulation; when GEN leans towards feminization, the pronunciation tends tolarger mouth opening, lighter nasal sound, flatter mouth shape, and front articulation. Therefore, besides being used to create derivative singers, the gender parameter can also be used to fine-tune details such as the singer's mouth shape and articulation.
- Portamento Timing (POR)
- Range: 0–127 Default: 64
- Controls theoffset timingof the pitch transition between two adjacent notes of different pitches (for example, from Do to Mi, the pitch does not change to Mi instantly, but rather the pitch transition is delayed or advanced slightly); the higher the value, the later the pitch change occurs. It actually acts on the beginning of the subsequent note, making it more convenient than using PIT to edit transitions.
- Some tutorials believe that the POR parameter changes the speed of pitch change (i.e., the duration required for the pitch transition), but in factthis is incorrect, a conclusion that can be drawn through the Pitch Rendering function (shortcut "P") in V4.
- The influence of POR on the pitch line is independent. PIT and PBS act directly on the area where the parameters are drawn and are not shifted by the POR parameter value.
- Pitch Bend (PIT / PB)
- Range: -8192 to +8191 Default: 0 (Supported by AI voicebanks)
- Controls pitch fluctuations and adjusts note pitch, functioning the same as the "Pitch bend wheel" on an electronic keyboard. The higher the value, the higher the actual output pitch compared to the note itself; conversely, the lower the value, the lower the pitch. A value of 0 means no pitch bend.
- Together with PBS, it determines pitch changes; these two parameters can be used in coordination to perform "sample zone switching" operations.
- Pitch Bend Sensitivity (PBS)
- Range: 1–24 Default: 2 (Supported by AI voicebanks)
- Controls the upper and lower limits of how the Pitch Bend (PIT) parameter affects pitch, measured in semitones. The higher the value, the larger the range. Together with PIT, it determines pitch changes but does not directly affect pitch itself.
- The formula for calculating pitch change is: PitchBend = PIT * PBS / 8192.
- Growl (GWL / GRW)
- Range: 0–127 Default: 0 (V4 and above only)
- Produces a growling effect in the singer's voice; the higher the value, the stronger the growling effect.Use with caution, except for newbies who want to turn their singer into Gurutamin.
- This parameter can only be called for voicebanks that include growl samples (V4 and V5 generations).
- Cross-Synthesis (XSY)
- Range: 0–127 Default: 0 (Only for multi-voicebank voicebanks in V4 and V6 6.5 and above)
- Adjusts the primary and secondary singer relationship of multi-voicebank voicebanks (Cross-Synthesis groups support some V4 and V3 sound sources; other voicebanks do not support it natively[10]). The higher the value, the closer the sound is to the secondary singer's voice; the primary and secondary singers can be changed in the singer settings.
- Airiness (AIR, Air)
- Range: 0–127 Default: 0 (V5 and above only; AI voicebanks V6 6.5 and above only)
- Similar to the function of BRE, setting it to maximum is like adding a _0 effect after the phoneme. This parameter can increase the proportion of airiness in the singing voice.
- Exciter (EXT, Exciter)
- Range: -64~+63 Default: 0 (V5 and above only)
- Adjusts the excitation value of the singer's voice, functioning similarly to an Exciter effect in general mixing.
VOCALOID Chronology
2000 | ||
|---|---|---|
| March | VOCALOID development begins. At the time, it was named "Daisy Project".[11] | |
2003 | ||
| February 26 | VOCALOID (1st generation) is announced. | |
2004 | ||
| January 15 | The first VOCALOID software "LEON」「LOLA" were released. | |
| July 1st | 「MIRIAM" were released. | |
| November 5th | 「MEIKO" were released. | |
2005 | ||
| June | The VOCALOID engine was upgraded to Ver1.1. | |
2006 | ||
| February 17 | 「KAITO" were released. | |
2007 | ||
| January | VOCALOID 2 was announced. | |
| June 29 | The first VOCALOID 2 voicebank "SWEET ANN" were released. | |
| August 31st | The first Japanese VOCALOID 2 voicebank "初音ミク" were released. | |
| December 27 | 「鏡音リン・レン" were released. | |
2008 | ||
| January 14 | 「PRIMA" were released. | |
| July 31st | 「GACKPOID" were released. | |
2009 | ||
| January 30 | 「巡音ルカ" were released. | |
| April 7 | NetVOCALOID service started. | |
| June 26 | 「Megpoid" was released. | |
| July 15 | 「SONIKA" were released. | |
| December 4th | 「氷山キヨテル」「歌愛ユキ」「SF-A2 開発コード miki" were released. | |
| December 22nd | 「BIG AL" download version went on sale. | |
2010 | ||
| February 25 | VOCALOID-flex announced to be available. | |
| April 29 | The game software "METAL GEAR SOLID PEACE WALKER", which uses NetVOCALOID and VOCALOID-flex, was released. | |
| April 30 | The expansion database for "Hatsune Miku", "Hatsune Miku APPEND", was released. | |
| July 13 | 「Tonio" were released. | |
| August 25 | 「Lily" was released. | |
| September 1 | 「VY1" were released. | |
| October 8 | 「GACHAPOID" was released. | |
| October 14 | VOCALOID-board announced. | |
| October 22 | 「猫村いろは" was released. | |
| December 8th | 「 歌手音ピコ" were released. | |
| December 13th | The iPad app "iVOCALOID VY1" and the iPhone/iPod touch app "iVOCALOID VY1t" (renamed to Vocalo Witter the following year) were released. | |
| December 27 | 「鏡音リン・レン" expansion database "鏡音リン・レンAPPEND" were released. | |
2011 | ||
| April 25 | 「VY2" were released. | |
| June 8 | VOCALOID 3 announced. | |
| October 21 | VOCALOID 3 released. | |
| 「VY1V3」「Mew」「Megpoid Power/Whisper/Adult/Sweet」「SeeU」released simultaneously. | ||
| December 16 | 「兎眠りおん" were released. | |
| December 21st | 「OLIVER" were released. | |
| December 22nd | 「CUL」「結月ゆかり" was released. | |
| December 23rd | 「Bruno」「Clara" were released. | |
2012 | ||
| January 27 | 「IA" were released. | |
| March 16 | 「Megpoid Native」 released. | |
| April 6th | 「蒼姫ラピス" were released. | |
| April 19 | "Lily V3" released. | |
| April 28 | 「galaco" announced. | |
| May 6 | 「MAYU" announced. | |
| July 12 | 「Luo Tianyi" were released. | |
| July 13 | 「GACKPOID NATIVE/POWER/WHISPER" released. | |
| December 5 | 「MAYU" were released. | |
| December 22nd | 「AVANNA" were released. | |
2013 | ||
| February 15 | "KAITO V3" released. | |
| June 20 | 「ZOLA PROJECT" released as a VOCALOID 10th anniversary commemorative product. | |
| July 11 | 「Yanhe" were released. | |
| August 31st | 「Hatsune Miku V3 English" released. | |
2014 | ||
| May 9th | 「v flower" were released. | |
| December 17 | VOCALOID4, "_VY1V4" were released. | |
2015 | ||
| February 4 | 「Cyber Diva" were released. | |
| February 10 | 「Xin Hua] (VOCALOID 3) released. | |
| March 18 | 「結月ゆかり" were released. | |
| March 19 | 「Megurine Luka V4X" released. | |
| July 16th | "v4 flower" released. | |
| July 17 | 「Yuezheng Ling] (VOCALOID 3) released. | |
| July 27th | 「Sachiko" were released. | |
| August 14 | 「Yuezheng Longya] passed the 100,000 repost challenge, was successfully approved, and the production of a VOCALOID 4 voicebank was announced. | |
| September 23 | 「ARSLOID" were released. | |
| December 24 | 「Kagamine Rin/Len V4X" released. | |
2016 | ||
| January 28 | 「Fukase" were released. | |
| February 20 | 「Stardust" were released. | |
| July 23rd | 「Luo Tianyi V4" design revealed, voicebank debut performance. | |
| July 30th | 「Otomachi Una" released | |
| August 31st | "Hatsune Miku V4X/V4 English" released. | |
| October 27 | 「Tohoku Zunko V4 Natural" released. | |
2017 | ||
| February 14 | 「Uni" were released. | |
| February 16 | 「兔眠りおん」「梦眠ネム" were released. | |
| May 10th | "Yuezheng Longya" released. | |
| June 3rd | 「LUMi" were released. | |
| July 12 | 「Masaoka Azuki」「Kobayashi Matcha" were released. | |
| August 12th | "Xin Hua V4 Japanese/Chinese" released. | |
| August 31st | "Hatsune Miku V4 Chinese" released. | |
| December 30 | "Luo Tianyi V4 Chinese" released. | |
2018 | ||
| April 26 | 「紲星 あかり" were released. | |
| May 21 | 「Luo Tianyi V4 Japanese」 released. | |
| May 30th | 「Zhiyu Moke" were released. | |
| June 14 | 「Mo Qingxian" were released. | |
| July 12 | VOCALOID5 announced. | |
| Voicebanks such as 「Amy, Chris, Kaori, Ken」 and 「VY1 V5, VY2 V5, CYBER DIVA II, CYBER SONGMAN II」 were released along with the editor. | ||
| 「Haruno Sora" were released. | ||
2019 | ||
| March 30 | 「Meika Hime & Mikoto" were released. | |
2022 | ||
| October 13 | VOCALOID6 released. | |
| Voicebanks 「SARAH, ALLEN, HARUKA, AKITO」 were released along with the editor. | ||
| 「Megpoid V6」 released. | ||
| October 14 | Pre-orders for 「Luo Tianyi V5」, 「Yanhe V5」, and 「Yuezheng Ling V5」 began. | |
2023 | ||
| February 12th | 「Luo Tianyi V5」, 「Yanhe V5」, and 「Yuezheng Ling V5」 released. | |
| March 7 | 「Po-uta" were released. | |
| May 9th | 「Fu Se" were released. | |
| June 20 | "ZOLA PROJECT V6" released. | |
| June 22 | "AI Otomachi Una" released. | |
| August 22 | "VX-β" demo released, and applications for irregular lottery use are open to Yamaha Music members. | |
| December 25th | "SAKURA, ASAHI, SHION, TAKU" released with the editor. | |
2024 | ||
| February 19 | "MICHELLE, LUCAS" released with the editor. | |
| April 18th | 「Hanahibi Koto" were released. | |
| April 24 | 「Shikiro En" were released. | |
| July 18 | 「GekiyakuV」「KazehikiV" were released. | |
| August 5th | "galaco BLACK/WHITE" released. | |
| September 30 | 「Tsuina-chan" were released. | |
| October 9 | "YUINA and NAOKI" were released with the editor. | |
| 「VOCALOID's Ci-chan" were released. | ||
| November 27 | 「Kanade" were released. | |
VOCALOID Version Overview
VOCALOID
Yamaha began developing VOCALOID in March 2000[12], and it was unveiled at the Musikmesse Frankfurt in Germany from March 5 to 9, 2003[13]. The first VOCALOIDs to be unveiledLeonandLolawere released by Zero-G on March 3, 2004, marketed as "Virtual Soul Vocalists." Leon and Lola were first exhibited at the NAMM Show on January 15, 2004[14]. Both were featured in Wired magazine and won the 2005 Electronic Musician Editor's Choice Award. Zero-G subsequently releasedMIRIAM, with the voice provided by Miriam Stockley. In late 2004,Crypton Future Mediaalso released its first VOCALOID productMEIKO. In June 2005, Yamaha upgraded the software engine to 1.1[15]. Yamaha subsequently released a patch to upgrade all VOCALOID software to 1.1.2, adding some new features, although the new features added after the upgrade varied between different software.[16]. A total of five VOCALOID voicebanks were released between 2004 and 2006. VOCALOID had no competitors in its early release, and its English version only had to compete with the subsequently released Cantor from VirSyn[17]. However, despite having Japanese voicebanks, the software lacked a Japanese interface; both the English and Japanese versions of the software featured an English interface, with the only differences being the logo and the color scheme of the main interface between versions. In 2011, version 1.1 was no longer supported by Yamaha and updates were discontinued[18]。
| Original English text |
|---|
|
Yamaha started development of VOCALOID in March 2000 and announced it for the first time at the German fair Musikmesse on March 5–9, 2003. The first VOCALOIDs, Leon and Lola, were released by the studio Zero-G on March 3, 2004, both of which were sold as a "Virtual Soul Vocalist". Leon and Lola made their first appearance at the NAMM Show on January 15, 2004. Leon and Lola were also demonstrated at the Zero-G Limited booth during Wired Nextfest and won the 2005 Electronic Musician Editor's Choice Award. Zero-G later released Miriam, with her voice provided by Miriam Stockley, in July 2004. Later that year, Crypton Future Media also released their first VOCALOID Meiko. In June 2005, Yamaha upgraded the engine version to 1.1. A patch was later released to update all VOCALOID engines to VOCALOID 1.1.2, adding new features to the software, although there were differences between the output results of the engine. A total of five VOCALOID products were released from 2004 to 2006. VOCALOID had no previous rival technology to contend with at the time of its release, with the English version only having to face the later release of VirSyn's Cantor software during its original run. Despite having Japanese phonetics, the interface lacked a Japanese version and both Japanese and English vocals had an English interface. The only differences between versions were the color and logo that changed per template. As of 2011, this version of the software is no longer supported by Yamaha and will no longer be updated. |
VOCALOID 2
VOCALOID 2 was released in 2007. Due to time constraints, VOCALOID 2 did not undergo public beta testing like previous versions; software updates were determined by issues reported by users.[19]The synthesis engine of this version andUI(Note)User Interfacewere both a massive innovation, and the Japanese version of the software featured a dedicated UI.[20]VOCALOID 2 implemented many new features, such as note auditioning, transparent control track, toggling between playback and rendering, and expression control[21]. Furthermore, human breath sounds and husky voices can be recorded into the voice library to create more realistic sounds[22]. VOCALOID 2 is not compatible with previous versions, and voice libraries from previous versions cannot be used in this version. In addition to PC software, NetVOCALOID services were also made public. The software was not localized, and software versions for each language could only have voicebanks in their respective languages, so althoughMegurine Lukapossessing an English voicebank, but the Japanese version of VOCALOID could only load its Japanese voicebank. VOCALOID had a total of 17 Japanese voicebank packages and 5 English voicebank packages, with a total of 35 voicebanks. Yamaha atY2 Autumn 2010 Digital Content Expo(Note)Y2 Autumn 2010 Digital Content Expodemonstrated VOCALOID 2 oniPhoneandiPadversions[23][24]. Subsequently, the iPhone and iPad versions of the software were released, using Yamaha's own VY1 voicebank.[25][26]
| Original English text |
|---|
|
VOCALOID 2 was announced in 2007. Due to time constraints, unlike the previous engine version, it did not have a public beta test and instead the software was updated as users reported issues with it. The synthesis engine and the user interface were completely revamped, with Japanese VOCALOIDs possessing a Japanese interface. New features such as note auditioning, transparent control track, toggling between playback and rendering, and expression control were implemented. One's breath noise and husky voice can be recorded into the library to make realistic sounds. This version is not backward compatible and its editor cannot load a library built for the previous version. Aside from the PC software, NetVOCALOID services are offered. Despite this, the software was not localized and VOCALOIDs of either English or Japanese would only possess that language version, so although Megurine Luka had an English library included, as a Japanese VOCALOID she only had access to the Japanese version of the software. In total, there were 17 packages produced for VOCALOID 2 in the Japanese version of the software and five in the English version; these packages offered 35 voicebanks between them in either English or Japanese. Yamaha announced a version of the VOCALOID 2 software for the iPhone and iPad, which exhibited at the Y2 Autumn 2010 Digital Content Expo in Japan. Later, this version of the software was released using the voice of Yamaha's own VOCALOID called VY1. |
VOCALOID 3
VOCALOID 3 was launched on October 21, 2011, along with several Japanese voicebanks and the first Korean voicebank. Yamaha provided ways for some V2 voicebanks to upgrade to V3[27]. V3 also included a system called Vocalistener (Bokarisu), which can automatically measure and adjust the speech synthesis parameters of a user's singing voice to synthesize more natural singing[28][29]. V3 supported more languages, including Chinese, Korean, and Spanish. V3 also supports using its own plugins to switch between standard and "classic(Note)classic" modes, the latter of which generates vocals that are less realistic than the former. Unlike previous products, V3's software and voicebanks are sold separately. The VOCALOID editor distributed with VOCALOID 3 voicebank software is only the "Tiny" version. Yamaha also grants licenses for plugins, allowing V3 to be used on more platforms, such as video games[30][31]. Additionally, the VOCALOID 3 engine supports triphone technology, which improves language compatibility compared to VOCALOID 2.[32]。
New technology in V3 was also applied to recreate the voice of singer Hitoshi Ueki, who passed away in 2007. This was the first time such technology was used to recreate a voice that had been lost. While this had been considered a possibility by some since the software's launch in 2004, it was only used for private purposes until 2011[33]。
| Original English text |
|---|
|
VOCALOID 3 launched on October 21, 2011, along with several products in Japanese and a Korean product, the first of its kind. Several studios are providing updates to allow VOCALOID 2 vocal libraries to come over to VOCALOID 3. It will also include the software "Vocalistener", which adjusts parameters iteratively from a user's singing to create natural synthesized singing. It will support additional languages including Chinese, Korean, and Spanish. It is also able to use plug-ins for the software itself and switch between normal and "classic" mode for less realistic vocal results. Unlike previous versions, the vocal libraries and main editing software are sold as two separate items. The vocal libraries themselves only contain a "tiny" version of the VOCALOID 3 editing software. Yamaha will also be granting the licensing of plug-ins and use of the VOCALOID software for additional mediums such as video games.Also, VOCALOID 3 has Triphone support unlike VOCALOID 2 which improves language capabilities. New technology is also being used to bring back the voice of the singer Hitoshi Ueki who died in 2007. This is the first attempt to bring back a singer whose voice had been lost, yet it had been considered a possibility since the software was first released in 2004. However, this is only being done for private use.[3 |
VOCALOID 4
In December 2014, VOCALOID 4 was officially released. Compared to V3, V4 has many new features, such as the pitch line and Pitch Snap Mode, and added two major parameters, GWL and XSY, further improving the editor's performance. In addition, users can upload their works to VOCALOID.NET. Besides the standalone VOCALOID 4 editor, "VOCALOID Editor for Cubase" was also released, enabling real-time note input via keyboard devices. After the initial release of V4, Yamaha also performed several updates, introducing more practical functions, such as quick switching of previously used parameters, quick switching between foreground and background parameters, quick execution of JOB plugins, batch erasing of lyrics, etc., and improved the stability of the program.
VOCALOID 5
On July 12, 2018, VOCALOID5 was released.
Compared to previous versions, V5 features over 1,000 phrases and over 1,000 audio samples, more than 100 singing style presets, 11 audio effects, and the editor supports VST as well as legacy V3 and V4 voicebanks.
Meanwhile, the project file format changed from VSQX (V3/V4) to VPR (V5); the editor still supports reading VSQX but no longer supports the earlier VSQ format.
This version also adjusted the original 12 major parameters to 13. It added support for Attack & Release effects, the ability to adjust the waveform of individual notes, real-time pitch rendering, and real-time input from MIDI instruments, with a maximum of 32 simultaneous tracks. Support for Job Plugins and Cross Synthesis (XSY) was removed.
For the voicebanks, four new voices were added, including VY1 (Japanese female), VY2 (Japanese male), CYBER SONGMAN II (English male), and CYBER DIVA II (English female).
- Amy (far left): English female voice, featuring a modern singing style, excellent technique, natural pronunciation, and a versatile voice. Suitable for genres from Tone and R&B to pop music.
- Chris (second from left): R&B English male voice with a wide range, gentle yet powerful lows, and expansive highs.
- Kaori (second from right): Japanese female voice with long sustain and an explosive singing style.
- Ken (far right): A bright, clear Japanese male voice, suitable for singing J-POP.
VOCALOID 6
Background: On September 3, 2019, Yamaha announced that with the help of NHK and many collaborators, they used voice materials from Hibari Misora to synthesize her singing voice through deep learning and created a 3D image to reproduce her performance. On that day, the deep learning-based VOCALOID:AI technology was announced.[34]
In March 2021, Yamaha Corporation registered the "VOCALOID:AI" trademark in mainland China.
The official version of VOCALOID 6 was released on October 13, 2022, featuring an AI synthesis engine, cross-lingual support, optimized rendering strategies, and Yamaha's latest VOCALO Changer (similar to an AI voice changer) function.
VX-β
On July 27, 2023, Yamaha Corporation released a teaser featuring only the letter "β" and the date August 8, 2023. On August 8, Yamaha announced the establishment of the limited-time project VOCALOID β-STUDIO to pursue the future of singing voice synthesis. The project is an experimental activity aimed at breaking and expanding the inherent impression of "singing voice synthesis" and establishing new norms in music production.
On August 22, 2023, VOCALOID β-STUDIO's first and only product, VX-β, was released. VX-β is a limited-time, limited-release DAW plugin; users applied and were granted activation eligibility through a lottery system. It can only be used withinDAW(Note)Digital Audio Workstation, allowing users to adjust the singer's tone quality and dynamics via knobs, and supporting the reading of vpr(V5,V6), ccs, svp, ust, and MusicXML formats. The first batch provided 9 voicebanks with the plugin, including prtv_0, prtv_1, prtv_2, prtv_3, Gazenβ, nagiβ, multiβ-N (including 12 female voicebanks f00-f11 and 5 male voicebanks m00-m04), Gekiyakuβ, and Kazehikiβ.
On September 14, 2023, VOCALOID β-STUDIO announced on Twitter the upcoming release of information for 4 new voicebanks. On September 26, 4 voicebanks based on VTuber characters—Kasukabe Tsumugi, Kanade Kanon, Anido Uge, and Ci-chan—were provided to all VX-β users.
The VOCALOID β-STUDIO project ended operations on March 31, 2024, and VX-β also ceased activation starting April 1, 2024.
On July 18, 2024, VOCALOID β-STUDIO posted that its official account would soon close, while a new version of VX-β would be released alongside VOCALOID 6, directly supporting the reading of VOCALOID 6 voicebanks without requiring secondary activation. On the same day, two third-party characters from VX-β, Gekiyaku and Kazehiki, joined VOCALOID 6 and released voicebanks; Yamaha also announced that the other four third-party characters from VX-β (Kasukabe Tsumugi, Kanade Kanon, Anido Uge, and Ci-chan) would have VOCALOID 6 voicebanks produced.
November 27, 2024[35], VX-β is provided to VOCALOID 6 users as a free bundle.
Piapro Studio Editor
In 2013, to provide users of its own VOCALOID voicebank products with a complete VOCALOID music production environment (DAW + voice editor + voicebank),Crypton Future Mediaauthorized by Yamaha to use the VOCALOID API, under the name ofPiapro(Crypton's creative platform), developed and released the VSTi version of the VOCALOID3 editor "Piapro Studio", inHatsune Miku V3、KAITO V3Bundled with voice bank software, the software supports both Windows and macOS. This software has a user interface and functions similar to the VOCALOID editor, can be launched as a VST plugin in a DAW, and allows for importing VOCALOID vocal sequence files (VSQ/VSQX) and MIDI files, as well as performing VOCALOID vocal editing (tuning).
Since December 2013, Crypton has enabled the ability to use other companies' voice bank software within Piapro Studio through software updates, improving the cost-performance ratio of Crypton's voice bank software.
When Crypton released its VOCALOID4 sound source series (namelyMegurine Luka V4X、Kagamine Rin/Len V4X、Hatsune Miku V4X), the editor was upgraded to a version supporting VOCALOID 4 (Piapro Studio v2.x) and natively (directly and easily without scripts) supports the E.V.E.C. technology provided by Crypton for V4X voice banks, allowing for detailed adjustments to the "expression" of the singing voice.[36]
The Hatsune Miku V4 Chinese (V4C) software includes the newly released Piapro Studio Standalone editor (i.e., the standalone version of Piapro Studio), which, as independent software, can replace the VOCALOID4 editor for VOCALOID vocal editing (supporting only the Hatsune Miku V4C voice bank).[37]
Given that the YAMAHA VOCALOID4 editor has long since ceased sales as of now (2026), Piapro Studio has become the most cost-effective VOCALOID3/4 vocal editor. Purchasing the cheapest Hatsune Miku V4 Chinese software (under 700 RMB) bundles the VSTi version of the Piapro Studio editor, which can edit various voice banks (the VOCALOID5 editor is priced at 27,500 JPY, approximately 1,300 RMB).
Crypton has not obtained the license to use the VOCALOID5/6 API, so Piapro Studio has no plans for compatibility with V5 or V6 voice banks. Piapro Studio NT is incompatible with VOCALOID voice banks. At the end of 2025, YAMAHA and Crypton began providing VOCALOID6 EDITOR LITE, a lightweight V6 editor, to some Crypton voice bank users.
VOCALOID Family
The VOCALOID family (commonly known as "V-family" on the Chinese internet) is a collective term for all singing voice synthesis software based on VOCALOID technology and their derived anthropomorphic characters. It includes, but is not limited to:MEIKO、KAITO、Hatsune Miku、Kagamine Rin/Len、Megurine Luka、Kamui Gakupo、GUMI、IA、Luo Tianyi、Stardustetc.
Others
TALKLOID
Also known as "intonation tuning," it refers to some producers using the sound sources in this software to generate results similar to human reading after tuningActually, there is still a bit of a singing componentin some works, which are usually used to create mini-theaters.
VOCALOID derivative
Please refer toTemplate:VOCALOID Sub-species。
Jinriki VOCALOID
Please refer toJinriki VOCALOIDentry. Please note that "Jinriki VOCALOID" is only a metaphorical term, and in reality, it has no relation to VOCALOIDno relation, and the current use of this name for this technologyDoes notrepresents the recognition of this term.
Notes
- ↑ [1]
- ↑ Crypton | VOCALOID 2 Special Feature
- ↑ The new version of VX-β was subsequently provided to VOCALOID 6 users as a free bundle on November 27, 2024.
- ↑ Compared to DYN64, the loudness of DYN36 is reduced by about 5dB, and the amplitude is about 56% of the original.
- ↑ This means the change in amplitude is linear. DYN0 means the amplitude is 0, the loudness is -∞dB, and no sound is emitted at all.
- ↑ This means the change in amplitude is exponential. The amplitude of DYN127 is 400% of DYN64, and the decibels increase by about 12dB.
- ↑ The changes in loudness and amplitude for BRI are much smaller than those for DYN, and the changes in both loudness and amplitude can be considered approximately linear.
- ↑ Compared to BRI64, the loudness of BRI0 decreases by about 12dB (amplitude is about 25% of the original).
- ↑ Compared to BRI64, the loudness of BRI127 only increases by about 0.72dB (amplitude is about 108.6% of the original).
- ↑ https://rsc-net.vocaloid.com/assets/pdf_files/FAQ/XSynth_V6.5_ENG.pdf
- ↑ A tribute to the world's first vocal synthesis track created by IBM employees using an IBM 7094 computer, "Daisy Bell". Later, to avoid copyright disputes, it was renamed "VOCALOID".
- ↑ Investigation and Research on Intellectual Property Rights of Digital Content: Evolving Content Business Models and Their Profitability and Legality - VOCALOID 2, Hatsune Miku, Users, UGM Websites, Rights Holders: Report
- ↑ "New Yamaha VOCALOID Singing Synthesis Software Generates Superb Vocals on a PC". Business Wire. AllBusiness.com. March 4, 2003. Retrieved October 25, 2010.
- ↑ "Computer Music Journal". MIT Press Journals. July 11, 2010.
- ↑ "Towards Even More Natural Virtual Vocals". DTM Magazine. June 16, 2005. November 14, 2009.
- ↑ Yamaha Corporation. Vocaloid. 2010-12-18. (Original contentArchived on 2011-08-24) (English).
- ↑ "VOCALOID Miriam" March 26, 2011
- ↑ Zero-G e-mail; "It is true that Yamaha are not supporting VOCALOID 1 - that means they will not make Leon, Lola or Miriam work 100% on the latest PC systems like Windows 7 or 64 bit. It is unlikely these voices will be upgraded to VOCALOID 3 or later versions. It could happen, but it is unlikely. It would be expensive and time consuming to re-work the phoneme databases, and at the moment time is better spent developing new voices."
- ↑ "2-6.html (VOCALOID 2 Information) Today is the last night... [Broken Link]". Crypton Future Media. July 12, 2007. May 13, 2012.
- ↑ "Interview of Hideki Kenmochi" (Japanese). DTM magazine (Terajima Joho Kikaku) 15 (2 (January Special ed. The VOCALOID CV01 Hatsune Miku)): pp. 36–39. January 2008.
- ↑ Yamaha. Key Features of VOCALOID2. VOCALOID official website. 2011-05-21. (Original contentArchived on 2011-08-28) (English).
- ↑ Inoue, Satoru (January 14, 2008). "Technology Frontier: Singing Synthesizer Gets Closer to a Natural Voice (Yamaha, Crypton Future Media, Animo)" (Japanese). Nikkei Business (Nikkei BP) (1424): pp. 78–80. ISSN 0029-0491.
- ↑ "Digital Content EXPO: VOCALOID on iPad/iPhone apps, Yamaha develops " . IT Media. October 14, 2010. Retrieved October 17, 2010.
- ↑ "Y2 Autumn 2010 |Digital Content Expo 2010". Digital Content Expo. Retrieved October 17, 2010.
- ↑ "iVOCALOID-VY1" . Apple Inc.. Retrieved December 13, 2010.
- ↑ "iVOCALOID-VY1t" Apple Inc.. December 13, 2010.
- ↑ 3-new-features-and-voicebanks/ "VOCALOID 3: Release Date, New Features and Voicebanks"[Broken Link]. VOCALOIDism. June 14, 2011. June 14, 2011
- ↑ Tomoyasu Nakano and Masataka Goto (July 6, 2009). "VocaListener: A Singing-to-Singing Synthesis System Based on Iterative Parameter Estimation". July 27, 2011
- ↑ Tomoyasu Nakano and Masataka Goto. "VocaListener: A System for Automatically Estimating Singing Voice Synthesis Parameters that Mimic the Singing Style of a User's Performance"July 9, 2010
- ↑ 歌声合成ソフト 「VOCALOID™」の新バージョン『VOCALOID™3』を開発. Yamaha. 2011-06-08. (Original contentArchived on 2017-11-30) (Japanese).
- ↑ "Yamaha Updates VOCALOID Vocal Music Synthesis Engine". Anime News Network. June 8, 2011. June 8, 2011.
- ↑ "Aoki Lapis and Megpoid Native Release Dates Announced". VOCALOIDsm. February 15, 2012. February 5, 2012
- ↑ "Speech Synthesizer Could ‘Resurrect’ Dead Singers". Wired. December 20, 2011. December 24, 2011
- ↑ 美空ひばりの新曲ライブの実現を支援 あの歌声を当社最新の歌声合成技術『VOCALOID:AI™』で再現 「NHKスペシャル AIでよみがえる美空ひばり（仮）」に技術協力. VOCALOID ( ボーカロイド・ボカロ ) 公式サイト. (Original contentArchived on 2022-07-09) (Japanese).
- ↑ https://x.com/vocaloid_yamaha/status/1861686397796597960
- ↑ 「巡音ルカV4X」紹介 第10回 日本語DEMO版＋Piapro Studio V2配布開始
- ↑ 可愛らしい声質で歌い上げる初音ミクの中国語バージョン「初音ミク V4 CHINESE」が、2017年9月上旬に登場します！
External Links
Official Related
- YAMAHA VOCALOID Official Website
- VOCALOID β-STUDIO Official Website
VOCALOID CHINA PROJECT(Operations ceased)
Voicebank Publishers
- Crypton Future Media, INC.
- Shanghai Henian Information Technology Co., Ltd.
- Internet Co,. Ltd
- AHS Co., Ltd.
- 1st PLACE
- EXIT TUNES(Now EMP)
- i-Style Project
- SBS Artech(Now SBS A&T)
- ZERO-G
Related Wiki Sites
- Fandom Wiki(English)
- Fandom English Lyrics Wiki(English)
- Hatsune Wiki on atwiki(Japanese)
- Chinese Lyrics Wiki on atwiki