This article is about the Synthesizer V Studio software known as a voice database. If you are looking for the Synthesizer V concept then click here.
History[]
Prior to release on Synthesizer V Studio, Saki was known as "JA-F2" by codename and later "Masaki" when released for Web Synthesizer V on December 31, 2019.[2]
2020[]
On June 26, 2020, Synthesizer V Studio made its debut and a livestream demonstrating the new engine was launched. Saki was featured in this livestream and was announced to be partnered with AH-Software Co. Ltd. in releasing her physical copies, as well as digitals from both them and Animen.[3] Pre-orders for Saki packages started on the same day.[4]
Trial versions of Saki were released for Web Synthesizer V on July 22.[5] She then released on July 30 as a Standard voice database.[6] Saki's Lite voice database was made available on the same day.
On October 30, an AI voice database was announced known as "Saki AI". Saki AI was released on December 25 and a Lite version was also made available.[7][8]
2021[]
On November 20, 2021, Saki Standard and AI were made available on Dreamtonics's official Taobao account, which they had opened as part of their new sales model for the Chinese market.
As of December 1, the Web version of Saki became non-functional as error messages were returned in spite of meeting minimum Chrome/Edge version requirements. The JA-F2 version became the only Web voice database to still work.
2022[]
On February 4, 2022, Saki Standard and AI were made available on the official Dreamtonics International Store. On June 13, Dreamtonics announced that they launched a new official store for the Chinese market built on WeChat's Mini Program. Users would be able to purchase the Synthesizer V Studio Pro editor and Dreamtonics' vocals through the application in a convenient manner, enjoying the automatic real-time distribution shopping experience. From June 13 to July 13, users were able to take advantage of unlimited 10% off coupons in the new store.[9] It was later clarified that the decision to open this new shopping option was made due to Taobao's new restrictions on the sales of virtual goods as of April 2022, especially for restrictions on shipments without logistics since virtual goods would need to be delivered via email.[10]
On June 30, it was announced that Saki would be receiving a physical standalone for her AI voice database, which featured a new design. This was expected to release on July 21, with pre-orders beginning on July 1.[11]
When users started to use Synthesizer V Studio Pro's 1.7.0 update (which also released on July 21 and introduced High Dynamics Voice Models (HDVM) and AI Retakes), they complained about unstable voice-quality results when re-rendering projects saved under earlier versions of the editor that made use of Saki AI (after updating her voice database to v123). On July 22, Dreamtonics explained that by default, the AI Retakes feature favors "rich timbre and expressions over smooth and stable vocalization", resulting in inconsistent behavior relative to previous versions;[12] as a workaround to stabilize the voice output, both Dreamtonics and AH-Software offered these steps below:[13][14]
- Empty the selection.
- Open AI Retakes panel, go to Timbre tab.
- Under Takes -> Global Settings, reduce Expressiveness.
Responding to requests from users, both Dreamtonics and AHS made available for download that same day rolled-back versions of Saki AI (as well as other affected AI voice databases distributed by them) by logging in to their respective websites.[15][16]
Updates[]
AI - Full Version[]
A free update for Saki AI was released following the Synthesizer V Studio 1.2.0 update.
Saki AI v110 - June 18, 2021[19]
This update supports the 3rd generation AI singing voice synthesis.
Saki AI v118 - December 23, 2021[20]
A free update for Saki AI was released following the Synthesizer V Studio 1.5.0 update. This enables her to support Cross-lingual Singing Synthesis.
Saki AI v119 - March 25, 2022[21]
A free update for Saki AI was released following the Synthesizer V Studio 1.6.0 update. This enables her to use the new Vocal Mode options. Four variations were included: "Chest", "Airy", "Open", and "Soft".
Saki AI v123 - July 21, 2022[22]
A free update for Saki AI was released following the Synthesizer V Studio 1.7.0 update. This enables her to use the new High Dynamics Voice Models and AI Retakes features. Her profile image was also replaced with the new pink version and the background illustration for the editor was removed.
Saki AI v124b1 - November 10, 2022[23]
A free update for Saki AI was released following the Synthesizer V Studio 1.8.0b1 update. This enables her to use the new Diffusion Probabilistic Models (DPM) feature.
Final version update compatible for Synthesizer V Studio 1.8.0 which enables Saki AI to use the new Diffusion Probabilistic Models (DPM) feature if the beta version was not installed previously.
A free update for Saki AI was released following the Synthesizer V Studio 1.9.0b1 update. This implemented the new rap feature available for the English and Mandarin languages. Cantonese Chinese was added to the Cross-lingual Singing Synthesis feature as well as general improvements to the overall function, resulting in more accurate pronunciation that sounds closer to a native speaker when singing foreign words.
A free update for Saki AI was released following the Synthesizer V Studio 1.9.0b2 update. This version of the vocal allowed for compatibility with the engine's second beta, which focused on rectifying several design issues identified by the user base in the 1.9.0b1 post-release survey.
Final version update compatible for Synthesizer V Studio 1.9.0 which enables Saki AI to use the rap feature (English and Mandarin Chinese only) and allows for Cantonese Chinese Cross-lingual Singing Synthesis if the beta versions were not installed previously.
A free update for Saki AI was released following the Synthesizer V Studio 1.10.0b1 update. This enhanced pitch generation by utilizing Reinforcement Learning from Human Feedback (RLHF), translating to fewer off-key errors, making better use of vibrato, and creating more expressive transitions between notes. This also allowed for the use of the new "Enhancement" slider in the AI Retakes panel, which controlled the level of changes introduced by RLHF. For Synthesizer V Studio Pro users, feedback submission was enabled, allowing users to consensually share their preferred takes with only Dreamtonics for research and development purposes, where models trained with this feedback would be deployed in future voice database updates. Users may also opt out of this feature if undesired.
A free update for Saki AI was released following the Synthesizer V Studio 1.10.0b2 update, which was the second beta version following version 1.10.0b1 that introduced enhancements to voice databases and integrated the latest user feedback via Reinforcement Learning from Human Feedback (RLHF). Regarding the editor, numerous reported issues were addressed for a more fluid experience. A new generation of pitch models were trained using the extensive feedback from users over the past month, and the new models offered greater accuracy in voice timbre reproduction, as well as a reduction in mispronunciations. Dreamtonics also focused on issues from the previous beta release that made Vocal Modes seem less distinct on certain vocals, ensuring clearer differentiation in the 1.10.0b2 update. The latest update also introduced a new feature in the Note Properties Panel that allowed for even division of a note into multiple syllables, proving to be a useful tool for songs with faster tempos.
Final version update compatible for Synthesizer V Studio 1.10.0 which enables Saki AI to use Reinforcement Learning from Human Feedback (RLHF) as well as other enhancements (as listed in v126b1 and v126b2's notes) if the beta version was not installed previously.
Saki AI v127 - October 26, 2023[41]
A free update for Saki AI was released that added support for the Japanese language to the rap feature.
Saki AI v128b1 - December 18, 2023[42]
A free update for Saki AI was released following the Synthesizer V Studio 1.11.0b2 update. This implemented support for the Spanish language in the Cross-lingual Singing Synthesis feature.
Final version update compatible for Synthesizer V Studio 1.11.0 which enables Saki AI to sing in Spanish through Cross-lingual Singing Synthesis if the beta version was not installed previously.
AI Lite[]
A free update for Saki AI Lite was released to coincide with the Synthesizer V Studio 1.3.0 update.
Voice Database Information[]
Demonstrations[]
Web Synthesizer V (JA-F2)[]
| Demonstrations |
|---|
|
| |
|---|---|
| 赤とんぼ (Akatombo) - Short Version |
Synthesizer V Studio Standard[]
| Demonstrations |
|---|
|
| |
|---|---|
| Ripples - Short Version | YouTube |
| 赤とんぼ (Akatombo) / カトレアの咲く頃に (Katorea no Saku Koro ni) | bilibili |
Synthesizer V AI[]
| Demonstrations |
|---|
|
| |
|---|---|
| 翼 (Tsubasa) / カトレアの咲く頃に (Katorea no Saku Koro ni) | YouTube |
| カトレアの咲く頃に (Katorea no Saku Koro ni) | YouTube bilibili |
| 翼 (Tsubasa) | YouTube bilibili |
| 赤とんぼ (Akatombo) / カトレアの咲く頃に (Katorea no Saku Koro ni) | bilibili |
| MASHIRA | Niconico YouTube |
| 脳汁どばどば (Noujiru Dobadoba) (2022 Remix) | Niconico YouTube |
| Dreamtonics Recommended Songs Playlist[47] |
|---|
|
| |
|---|---|
| 鳳仙花 (Housenka) | Niconico YouTube bilibili |
Voice Databases[]
Product Information
Optimum
Range: C4
Versions
Lite Version/FLT Available?: No Starter Pack Available?: No Web Version Available?: Yes
Package details as noted:
JA-F2 (later what came to be "Saki") is a young feminine voice database based on a professional singer. She was made to demonstrate the capabilities of Web Synthesizer V and was the first to be made available for it. This minimized voice database was optimized for the current trends in the audio production industry.[48]
Phonetic notes as noted:
Phoneme format: Romaji
Software issues as noted:
- JA-F2's voice database was released after Synthesizer V and prior to Synthesizer V Studio, which was before the commercial versions (through Saki) were made available. Web Synthesizer V was noted to be more of a "branch off" of the Synthesizer V Studio development.[49] Therefore, it should be noted that JA-F2 may be very different to the Saki voice databases.
- As Web voice databases are generally restricted to monopitch (one pitch) only, JA-F2's range is noted to be C4 in this version.
System requirements as noted:
Supported web browsers:
- Google Chrome (version 77 and up)
- Microsoft Edge (Chromium-based, version 79 and up)
- Mozilla Firefox (nightly builds only)
Voice database sample
JA-F2 Solfège:
Product Information
Optimum
Genre: Ballads, Pop, & Rock Tempo: 80 ~ 150 BPM Range: F3 ~ E5 Pitch Group: G#3, C4, F4, B4 (Falsetto)
Versions
Lite Version/FLT Available?: Yes (Lite) Starter Pack Available?: Yes Web Version Available?: Defunct
Package details as noted:
Saki is a young feminine voice database that was ideally marketed toward professional musicians. According to AH-Software Co. Ltd., she has a natural singing voice that adapts to any occasion. She was described to work well in the high ranges.
Vocal traits as noted:
AH-Software noted that Saki is particularly good at ballads and pop, but also wrote that she can handle rock in her higher registers.
Phonetic notes as noted:
Phoneme format: Romaji
Software issues as noted:
- As the Lite and Web voice databases are generally restricted to monopitch (one pitch) only, Saki's range is noted to be C4 in these versions.
System requirements as noted:
- OS: Windows 10/8.1; Mac OS X: 10.11 and up; Linux Ubuntu 16.04 and up.[50] In or around October 2021, the AH-Software product page changed the Linux requirement to Ubuntu 18.04 or later (64-bit).[51]
- In November 2021, with the release of Windows 11 and the retirement of 32-bit support, AHS reworded the OS spec to: Windows 11/10/8.1 or later (64-bit); Mac OS X: 10.11 or later; Ubuntu 18.04 or later (64-bit).[52]
- CPU: Intel Core i3 or higher, or equivalent AMD processor.[50] AHS later increased this to: Intel Core i5 or higher, or equivalent AMD processor.[51]
- Memory: 2GB or more.
- HDD: 1GB or more for one voice database; 2GB or more if installing both Standard and AI databases.
- Display resolution: 1280x800 or higher resolution.
- Others: DVD-ROM drive (physical version only), audio device, internet connection.
License agreements as noted:
- Dreamtonics Co., Ltd.'s End User License Agreement - Japanese
- Dreamtonics Co., Ltd.'s End User License Agreement - English
- Dreamtonics Co., Ltd.'s End User License Agreement - Simplified Chinese
- AH-Software's End User License Agreement - Japanese
- AH-Software's End User License Agreement - English
- Regarding the Lite version:
- AH-Software does not accept technical inquiries regarding free software nor do they guarantee or support the operation of all free software.
- AH-Software is not liable for any damages (including loss of data) caused by using free software. Users are cautioned to use the free software at their own risk.
- The program may not be transferred, sold, lent, distributed or rented without AH-Software's permission, regardless of whether it is paid or free of charge.
- Reverse engineering, including modification, decompilation, and disassembly, of their program for any purpose is prohibited.
- Saki Standard Lite may not be used commercially in any way
- When using the Lite version of this voice database, "This work uses the Lite version" must be mentioned on any uploads.
Voice database sample
- Saki Standard Solfège:
- Saki Standard Lite Solfège:
- A preview of Saki's Synthesizer V Studio Standard voice database (provided by the Dreamtonics website):
Product Information
Optimum
Genre: Ballads, Pop, & Rock Tempo: 80 ~ 150 BPM Range: B3 ~ A#4[53]
Versions
Lite Version/FLT Available?: Yes (Lite) Starter Pack Available?: Yes Web Version Available?: No
Package details as noted:
Saki is a young feminine voice database that was ideally marketed toward professional musicians. According to AH-Software Co. Ltd., she has a natural singing voice that adapts to any occasion. She was described to work well in the high ranges. According to Dreamtonics, Saki's AI was built for maximal realism.
Vocal traits as noted:
- As per the Synthesizer V Studio 1.7.0 update launched on July 21, 2022, through High Dynamics Voice Models (HDVM), which later migrated to Diffusion Probabilistic Models (DPM) per the 1.8.0b1 update on November 10,[54] AI voice databases may use AI Retakes to create multiple "takes" of a sung section with variations in pitch, dynamics, and timbre. This replicates the randomness of human vocalists, adds a more realistic feel, and allows users to find an ideal vocal without the need for "laborious tuning." This feature is only available in the Synthesizer V Studio Pro Editor.[55][56]
- This does not apply to the Lite version.
Phonetic notes as noted:
- Saki AI's pitch model was noted to be trained on 5 songs.[57]
- Phoneme format: Romaji
- On December 23, 2021, with the announcement of Synthesizer V Studio 1.5.0, all AI voice databases would be able to use Cross-lingual Singing Synthesis. This allows Saki AI to be able to sing in not only Japanese but also in English and Mandarin Chinese using this new feature.[58] Since the release of her v125b1 update on April 18, 2023, she can also sing in Cantonese Chinese with this feature; since her v128b1 update on December 18, she can also sing in Spanish with this feature. Cross-lingual Singing Synthesis is only available in the Synthesizer V Studio Pro Editor.
- This does not apply to the Lite version, thus restricting the voice database to Japanese only.
Vocal Modes as noted (Synthesizer V Studio Pro Editor only):
- As of March 25, 2022, the paid version of Saki AI became compatible with the Vocal Mode function introduced in Synthesizer V Studio 1.6.0. Her v119 update included four variations:
- "Chest"
- "Airy"
- "Open"
- "Soft"
Software issues as noted:
- On February 19, 2021, Dreamtonics noted that users were experiencing issues with Synthesizer V Studio crashing along with the loss of activation statuses after upgrading to Saki AI v104.[59] It was noted that upgrading to Synthesizer V Studio 1.2.1 and re-entering the activation code would solve this issue[60]
System requirements as noted:
- OS: Windows 10/8.1; Mac OS X: 10.11 and up; Linux Ubuntu 16.04 and up.[61] In or around October 2021, the AH-Software product page increased the Linux requirement to Ubuntu 18.04 or later (64-bit).[51]
- In November 2021, with the release of Windows 11 and the retirement of 32-bit support, AHS reworded the OS spec to: Windows 11/10/8.1 or later (64-bit); Mac OS X: 10.11 or later; Ubuntu 18.04 or later (64-bit).[52]
- CPU: Intel Core i3 or higher, or equivalent AMD processor.[61] AHS later increased this to: Intel Core i5 or higher, or equivalent AMD processor.[51]
- For high-quality editing of AI audio: Intel Core series 4th generation i5 (i5-4xxx) or higher is recommended, or AMD Athlon X4 845 or higher, including Ryzen series.[62]
- Memory: 2GB or more.
- HDD: 1GB or more for one voice database; 2GB or more if installing both Standard and AI databases.
- Display resolution: 1280x800 or higher resolution.
- Others: audio device, internet connection.
License agreements as noted:
- Dreamtonics Co., Ltd.'s End User License Agreement - Japanese
- Dreamtonics Co., Ltd.'s End User License Agreement - English
- Dreamtonics Co., Ltd.'s End User License Agreement - Simplified Chinese
- AH-Software's End User License Agreement - Japanese
- AH-Software's End User License Agreement - English
- Regarding the Lite version:
- AH-Software does not accept technical inquiries regarding free software nor do they guarantee or support the operation of all free software.
- AH-Software is not liable for any damages (including loss of data) caused by using free software. Users are cautioned to use the free software at their own risk.
- The program may not be transferred, sold, lent, distributed or rented without AH-Software's permission, regardless of whether it is paid or free of charge.
- Reverse engineering, including modification, decompilation, and disassembly, of their program for any purpose is prohibited.
- Saki AI Lite may not be used commercially in any way
- When using the Lite version of this voice database, "This work uses the Lite version" must be mentioned on any uploads.
Voice database sample
- Saki AI Solfège:
- Vocal Mode Solfèges:
- "Chest":
- "Airy":
- "Open":
- "Soft":
- "Chest":
- Vocal Mode Solfèges:
- Saki AI Lite Solfège:
References[]
- ↑ In combination with the Synthesizer V Studio Pro editor, GENBU R2, and AiKO R2
- ↑ https://twitter.com/khuasw/status/1212025665823723522
- ↑ https://www.youtube.com/watch?v=rQNVX90g1C4
- ↑ https://twitter.com/ahsoft/status/1276351253073629185
- ↑ https://twitter.com/ahsoft/status/1285822368712032257
- ↑ https://twitter.com/ahsoft/status/1276349537079029760
- ↑ https://twitter.com/ahsoft/status/1322055701938622464
- ↑ https://www.ah-soft.com/press/synth-v/20201030.html
- ↑ https://t.bilibili.com/671200609838301200
- ↑ https://www.bilibili.com/read/cv17082866
- ↑ https://twitter.com/ahsoft/status/1542711016860962816
- ↑ https://twitter.com/dreamtonics_en/status/1550520904291012609
- ↑ https://twitter.com/dreamtonics_en/status/1550520905868009472
- ↑ https://twitter.com/ahsoft/status/1550385484027035649
- ↑ https://twitter.com/dreamtonics_en/status/1550520907159851009
- ↑ https://twitter.com/ahsoft/status/1550385686389616640
- ↑ https://twitter.com/dreamtonics_en/status/1362598004100800519
- ↑ https://www.youtube.com/watch?v=D_cXLpO0qIw
- ↑ https://twitter.com/ahsoft/status/1405724392936509440
- ↑ https://dreamtonics.com/en/cross-lingual-support-for-synthesizer-v-ai/
- ↑ https://dreamtonics.com/en/synthesizer-v-studio-1-6-0-update/
- ↑ https://dreamtonics.com/en/synthesizer-v-studio-1-7-0-update-ai-retakes-feature/
- ↑ https://dreamtonics.com/en/synthesizer-v-studio-1-8-0b1-update/
- ↑ https://dreamtonics.com/en/synthesizer-v-studio-1-8-0-final-update/
- ↑ https://twitter.com/ahsoft/status/1595674717200916481/
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-9-0b1-update-rap-cantonese-and-more/
- ↑ https://twitter.com/dreamtonics_en/status/1648235081608343552
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-9-0b2-update/
- ↑ https://twitter.com/dreamtonics_en/status/1660888307482656771
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-9-0-final-update/
- ↑ https://twitter.com/dreamtonics_en/status/1671427830629154816
- ↑ https://twitter.com/dreamtonics_en/status/1686648078571655168
- ↑ https://t.bilibili.com/825177687051993121
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-10-0b1-update-enhancing-pitch-generation-with-user-feedback/
- ↑ https://twitter.com/ahsoft/status/1686653272474738689
- ↑ https://twitter.com/dreamtonics_en/status/1707304150059639108
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-10-0b2-update/
- ↑ https://twitter.com/ahsoft/status/1707304906535956719
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-10-0-final-update/
- ↑ https://twitter.com/ahsoft/status/1714919114182181140/photo/1
- ↑ https://twitter.com/ahsoft/status/1717451110305366047
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-11-0b2-update/
- ↑ https://dreamtonics.com/synthesizer-v-studio-1-11-0-update/
- ↑ https://twitter.com/ahsoft/status/1775800933538320621/photo/4
- ↑ https://twitter.com/ahsoft/status/1405724076769910792
- ↑ https://twitter.com/ahsoft/status/1405725227305213955
- ↑ https://space.bilibili.com/593592382/favlist?fid=1655788882 - These songs were featured in an official bilibili playlist featuring demos and recommended songs. Certain users were invited to create songs using earlier versions of the voice database and release the songs in sequence after the product's release as a complement to official demo tracks. Although this did not apply to Saki when she first released, she may be present as a main vocalist in examples featuring the other Dreamtonics products, which would be added to this demo table. It should also be noted that certain songs may not be involved with the early invitation and were added based on the company's preference.
- ↑ https://twitter.com/khuasw/status/1212028889033297921
- ↑ https://twitter.com/khuasw/status/1212027759171534848
- ↑ 50.0 50.1 AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived June 26, 2020
- ↑ 51.0 51.1 51.2 51.3 AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived October 21, 2021
- ↑ 52.0 52.1 AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived November 26, 2021
- ↑ https://img.alicdn.com/imgextra/i1/54187543/O1CN01cLoZuW25akRabm5si_!!54187543.png - Initially, Saki AI's range was not officially listed. The only range that was available at the time was "F3 ~ E5" presented on ANiCUTE's product page, however, the information was taken from Saki Standard. An official range was not given until March 15, 2024 through Dreamtonics's Taobao product page.
- ↑ https://dreamtonics.com/en/synthesizer-v-studio-1-8-0-final-update/
- ↑ https://twitter.com/dreamtonics_en/status/1550048543624564736
- ↑ https://twitter.com/dreamtonics_en/status/1550048543624564736
- ↑ https://twitter.com/khuasw/status/1343127241375457280
- ↑ https://twitter.com/dreamtonics_en/status/1473941453810577410
- ↑ https://twitter.com/dreamtonics_en/status/1362607750782414848
- ↑ https://twitter.com/dreamtonics_en/status/1362709006544826370
- ↑ 61.0 61.1 AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived December 30, 2020
- ↑ AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived May 13, 2021