KVR Audio

Pelicanomicon · Post by **Pelicanomicon** » Sat Jun 12, 2021 6:11 pm

VOLOR has announced Eleanor Forte AI for 2021!
https://twitter.com/OfficialVolor/statu ... 8670590982

Reminder to people who bought AHS Saki, you can get a free copy of Saki AI until 00:00 AM June 18, 2021 JST
https://auth.dreamtonics.com/keymatchev ... 20Campaign

Since this is a new page: Eclipsed Sounds is crowdfunding a native English voicebank for Synthesizer V Studio based on the voice of stage singer Emma Rowley
https://igg.me/at/synth-v-solaris

Pelicanomicon · Post by **Pelicanomicon** » Sun Jun 13, 2021 6:26 am

Eclipsed Sounds held a second livestream on their crowdfund for a native English AI voicebank for Synthesizer V Studio, ending on June 28 2021. They have reached $20,000 of their $40,000 goal.

Timestamps for points KVRists might be interested in:
https://www.youtube.com/watch?v=ab4AQRLxaX4

14:40/19:35 Do you have a breakdown of how the required funding will be used?
(NDA with Dreamtonics about Synth V development and licensing firmly bars them from giving a breakdown. "What we can tell you is that the four of us are making absolutely no profit off of this and all of it is going toward exactly what it says in the crowdfund"--toward demo musicians, Dreamtonics, Emma Rowley for voicing SOLARIS, merch artists, Indiegogo fees, etc.)
22:27 Will copies need to wait until the boxed version comes out to use SOLARIS, or will we recieve a digital copy as well?
(Codes will go out by email 1 week before full release to digital and physical backers alike.)
25:06 Example of singing samples for training SOLARIS' AI and thus what they want the final product to sound like: "A clear and powerful vocal."
27:30 How long it took for SOLARIS [the test bank] to be recorded?
28:19 2 new tiers have been added
30:30 What would be the plan for Eclipsed Sounds' future as a company if SOLARIS is unable to be funded?
(Talks about other funding avenues, what if it is impossible to fund at all. Also, "Given our current resources--looking at our own bank accounts and small business loans as an option--we are closer than it seems to being able to make SOLARIS. [...] It is just necessary to get to a certain threshold and then we'll be able to make up the rest ourselves.")
32:45 How much softer will we be able to go? [...] Not volume, but tone.
34:15 Does the AI version have a different sample set than the normal one? Just wondering if it sounds different.
38:25 If funding for SOLARIS doesn't reach its goal, would other synths (for example Maghni AI* if it reaches completion) be considered?
39:20 Mentions possibility of PayPal donation link for people who want to contribute but can't use Indiegogo's payment methods
41:20 Given the lack of male English voicebanks, how does [Emma Rowley's] voice translate via the Masculine parameter?
44:58 What happens when you adjust the Tension parameter, either with AI or Standard pitch?
47:32 What would the legal situation be for the commercial use of SOLARIS original songs or use of the character?
48:48 In the future, will you all put on a voice contest for choosing the next voice actor or voice provider for future voicebanks?
(They talk about the merit of choosing a professional to train an AI.)
53:46 How to swap to a new crowdfund tier
55:03 Are there any vocal types you all feel are missing right now in the English space?
(They hope to aim beyond pop with Emma Rowley, Rowley also sings jazz. One dev talks about a "true rock vocal.")
57:19 Regarding use of the character does that include the official art, or is that strictly copyrighted?
1:00:21 Circus and GHOST mentioned for demo songs
1:06:20 If you have so many people making demos already, is it still viable for smaller producers to send in your portfolios?
(Yes, still looking for demos and for future projects.)
1:08:34 Vocal synths have struggled in the Western music community. Do we have any ideas on how to market her to the outside of the vocal synth community?
(They mention realism and Synthesizer V's realism--esp with AI vocals--as a possible selling point in the West. "We're also working on some plans on future marketing.")
1:12:10 Has your team been advertising the crowdfund on sites other than Twitter or TikTok?
(They're on other social media. Strict rules on music-making forums prevent them from self-advertising.)
1:15:06 Where will SOLARIS be sold when she's finished?
(Directly on Eclipsed Sounds' own website. Maybe with a distributor, still working on it.)
1:54:16 Post an illustration of SOLARIS and win a digital copy of Synthesizer V Studio Pro
1:57:16 Physical box mockup

*Maghni AI is a potential Western vocal synthesizer from another startup, Misbah Studios. The team includes developers with experience on English-language Vocaloids like PowerFX's V3 Oliver. It was initially announced for Kickstarter crowdfunding in December 2020.
https://twitter.com/VocaTone/status/1342420884838096901
The crowdfund was postponed shortly after as vocal synth fans were expressing concern about donating without proof of even alpha software or voice provider/synthesized examples. Devs are currently working on a demo.

Pelicanomicon · Post by **Pelicanomicon** » Fri Jun 18, 2021 6:08 am

So much news omg

https://twitter.com/dreamtonics_en/stat ... 8396389378

Dreamtonics has updated Synthesizer V Studio to 1.3.0 with Gen 3 AI improvements to singing at higher pitches and contextual breath sounds. Watch to the end, they also teased a male English AI voice!

https://www.youtube.com/watch?v=ir3eb9lwKp8

All AI voices incl. Lite versions have been simultaneously updated.
Dreamtonics link to free Synthesizer V Studio Basic editor and free voices incl. Tsurumaki Maki English AI Lite:
https://resource.dreamtonics.com/download/

AHS' Tsurumaki Maki English Standard and AI are officially on sale

Manufacturer download, checkout in your lang/currency, guest checkout: https://www.ah-soft.com/product/download.html

Products names on the store page:

Synthesizer V Studio Pro ~ JPY ¥9,800
Synthesizer V 弦巻マキ English ~ JPY ¥9,680
Synthesizer V 弦巻マキ English AI ~ JPY ¥9,680

Eclipsed Sounds' crowdfund for a native female English AI voicebank based on a professional singer has reached 57% with 10 days to go:
https://igg.me/at/synth-v-solaris

At the request of people who can't pay thru Indiegogo, they've opened up direct Paypal pledges. You can ask for the Indiegogo perks like a digital copy of the completed voicebank:
https://twitter.com/eclipsedsounds/stat ... 5806403585

Dreamtonics once again endorsed Eclipsed Sounds' crowdfund:
https://twitter.com/dreamtonics_en/stat ... 3183636489

Pelicanomicon · Post by **Pelicanomicon** » Sun Jun 20, 2021 11:58 pm

Eclipsed Sounds' campaign to make a native English AI voice got a big boost to 69% funded on Indiegogo today--70% including Paypal direct pledges. Probably in no small part to Vocaloid producer Circus' June 18 livestream of working on a demo song for SOLARIS in FL Studio:
https://twitter.com/VocaCircus/status/1 ... 2657704961

7 days to go until the crowdfund ends!
https://igg.me/at/synth-v-solaris

Eleanor users, VOLOR has teased a membership program. In the same thread, they say "We are going to loosen the profit restrictions on VOLOR's SynthV products." From the way it's tweeted it's not clear (to me, anyway) if you have to join for less restrictions, but details are TBA:
https://twitter.com/OfficialVolor/statu ... 3188480001

Pelicanomicon · Post by **Pelicanomicon** » Thu Jun 24, 2021 6:43 pm

With 4 days to go, Eclipsed Sounds has reached 86% of their Indiegogo crowdfund to create a native English AI voice for Synthesizer V...BUT, with outside donations and a generous private investment, they have already exceeded their initial goal of $40,000.

SOLARIS is now fully funded

Pelicanomicon · Post by **Pelicanomicon** » Sat Jun 26, 2021 10:02 pm

Holy crap, that was fast. Anicute will be opening the first round of preorders for VOLOR's Eleanor Forte AI from July 3 to July 31:
https://twitter.com/OfficialVolor/statu ... 7100109832

1st round preorder price is 20% off (~$70) with additional discount for previous Anicute customers. 1st round buyers will get early access before release.

They will also offer a bundle with Synthesizer V Studio Pro (~$148) and you can get the synth before Eleanor arrives:
https://twitter.com/OfficialVolor/statu ... 8429720580

Projected release is 2021 Q4.

SOLARIS is now at 93% of the Indiegogo crowdfund goal with a little less than 1 1/2 days left.

Pelicanomicon · Post by **Pelicanomicon** » Tue Jun 29, 2021 7:41 am

https://www.indiegogo.com/projects/new- ... pdates/all

On June 28, Eclipsed Sounds closed their crowdfund for a professionally performed, native English AI voicebank by exceeding their Indiegogo goal

The end total with outside donations was $50,417.93, so GJ them for running a successful campaign that focused on social media and fans.

They have also changed the name of the project from SOLARIS to SOLARIA. The projected digital release is in January 2022.

Pelicanomicon · Post by **Pelicanomicon** » Sat Oct 16, 2021 10:38 pm

New news

ANRI is a new AI female English voicebank from AUDIOLOGIE

https://soundcloud.com/audiologie_en/p2l

AUDIOLOGIE is "a United States based virtual vocalist production company, specialising in voice databases for Synthesizer V Studio." Crusher (a well-known Western Vocaloid producer) is a member.

Digital preorders are open on their website until 10th December 2021 at 23:59 EST, goes on sale 11 December 2021. Per their "About Us" page, Lite version planned for December 18, 2021.

https://audiologie.us/
US$ 79.99 sale, reg. 89.99

Or get bundles thru their partner VOLOR/AniCute
https://www.anicute.com/pages/anri%20_ai_pre-order

As for Eleanor Forte AI, VOLOR emailed the following to early bird buyers around a week ago:

Regarding Eleanor Forte AI's progress, we expect that the database will be completed by the end of October, meaning VOLOR can distribute the database to first-round customers in November 2021. We were pleased to hear from Dreamtonics that the recording quality of the database is excellent.

Discounted preorders for Eleanor Forte AI and bundles at
https://www.anicute.com/pages/early_bird_discount
NT $2,190 / ~ US $78 (a bit over 10% off)

For those tracking Eclipsed Sounds' crowdfunded SOLARIA voicebank, projected release in January 2022
https://www.indiegogo.com/projects/new- ... pdates/all

28 June 2021: Campaign ended
29 July 2021: Indiegogo released the funds
29 August 2021: Data collection finished, currently processing recordings
7 October 2021: Nearing completion of audio clean-up phase

As always, you don't need to buy SVS Pro to use any paid voicebanks. There is a freeware Basic edition but you will be locked out of useful features like VST/AU and alternate phoneme recordings.

Dreamtonics' Synthesizer V Studio Basic download + free Lite voices for English/Chinese/Japanese:

https://resource.dreamtonics.com/download/

Pelicanomicon · Post by **Pelicanomicon** » Sun Oct 17, 2021 12:02 am

Misc points of interest re Dreamtonics/Synthesizer V development

Dreamtonics is working on a text-to-speech synthesizer
https://twitter.com/dreamtonics_en/stat ... 9766682629

Dreamtonics shows off demo for new unnamed female Mandarin AI voice
https://www.bilibili.com/video/BV1eh411J74T

Quadimension is developing Xingchen (aka "Stardust") for Synth V.
https://twitter.com/SynthVNews/status/1 ... 0121715722
A Xingchen voicebank was previously sold for VOCALOID4. There is a trend of notable Vocaloid "characters" migrating to other companies' engines (Crypton Future Media's Hatsune Miku and friends to Piapro Studio NT; Vocalomakets' Yuzuki Yukari, 1st Place's IA and SSS's Tohoku Zunko to CEVIO AI)

A crowdfund for a new female Jpn Synth V AI vocal hit its initial goal of ¥6,800,000 on launch day 8 October 2021 and already exceeded its final stretch goal of ¥10,000,000
https://ubgoe.com/projects/68

Pelicanomicon · Post by **Pelicanomicon** » Tue Nov 23, 2021 7:52 am

If you preordered Eleanor Forte AI from AniCute back in July, check your email for early bird access! Release date for regular customers coming soon.

This demo from Neutrogic showcases her default sound:
https://soundcloud.com/giraffe-kun/synt ... icial-demo

You must update to Synthesizer V Studio 1.4.0 to use Eleanor AI. Changelog is here:
https://dreamtonics.com/en/synthesizer- ... -0-update/

New features:
Auto-Process: regroup auto pitch tuning related features under “Auto-Process” menu.

Auto-Process: add support for 4th generation pitch models.

Auto-Process: add a customizable version of Auto Pitch Tuning independent of the voice database. (only for Pro edition)

Auto-Process: add “Ornament Selected Notes” command that automatically splits the notes based on the melody context. (only for Pro edition)

Voice panel: add “Use relaxed consonants” option for phoneme conversion rules that better suit American English.

Lyrics Processing: new lyrics shorthand that inserts a glottal stop symbol (“cl”) for notes starting with a single quote

Pelicanomicon · Post by **Pelicanomicon** » Sun Nov 28, 2021 8:03 pm

VOLOR has announced that Eleanor Forte AI will officially be released on 6 December 2021

https://twitter.com/OfficialVolor/statu ... 9122102275

They also made the free Lite version of Eleanor AI available

Direct link: https://drive.google.com/drive/folders/ ... ocp31ED2Vr

Discounted preorders still up at AniCute, VOLOR's storefront
NT $2,450 / ~ US $88
Final price looks to be NT $2,690 / ~ US $97
https://www.anicute.com/

Other news:
Eclipsed Sounds posted that SOLARIA demos should go up in December. They're on track to release to crowdfund backers in January 2022, with official product going on sale mid-January.
https://www.indiegogo.com/projects/new- ... pdates/all

Link to freeware Synthesizer V Studio Basic, reduced features, works with Lite and paid voices
https://resource.dreamtonics.com/download/

Spitfire31 · Post by **Spitfire31** » Sun Nov 28, 2021 10:02 pm

I don't think it's an exaggeration to suggest that Eleanor Forte AI in Synthesizer V Studio Pro 1.4 have moved the goalposts when it comes to 'digital singing'. Just listen to this piece:
https://www.youtube.com/watch?v=46cDzEBZ274

User 555 · Post by **User 555** » Mon Dec 06, 2021 11:22 pm

Eleanor Forte AI is released today. Got a download link today.

Pelicanomicon · Post by **Pelicanomicon** » Tue Dec 07, 2021 8:54 am

So many voices debuting close together, feels like the sun goddess really lit a fire under developers' ?!@#*s.

I'm quite satisfied with Eleanor AI's audio quality--that was my biggest worry as the voicer's identity is undisclosed and we don't know what recording facilities they have access to. Install size on Win is less than 26MB, it is crazy how something that tiny sounds so good.

And I really like how the flagship Dreamtonics voices are natural-sounding and mature. Here's an example of their new AI female Chinese voice mentioned above, now named "Qingsu". The song was originally composed for VOCALOID Luo Tianyi,, it is 勾指起誓 ("Pinky Swear") by ilem:

https://www.youtube.com/watch?v=jjEmSDUaF4o
Meanwhile, AUDIOLOGIE's ANRI will release this week on 11 December 2021. If you preordered Eleanor AI from AniCute, you can log in to get ANRI for a few dollars off.

AUDIOLOGIE reposted ANRI's "Programmed 2 Love" demo using the final version of the voicebank with SVS's 4th gen AI:

https://soundcloud.com/audiologie_en/sy ... n-remaster

Pronunciation is excellent but she sounds kinda grainy compared to Eleanor...

I do have some small gripes about SVS:

The UI is very mouse-centric. Awkward for pen users, you can't horizonal zoom without the mouse wheel.
Sparse documentation. There's an official guide for newbies, but it's in Japanese Not great for expanding your Western userbase. The online manual is for Gen 1 Synthesizer V and still lists removed features like "Glottal Effects" and "Phoneme Viewer"!
- BTW, Eleanor AI has an undocumented [br] breath phoneme.
VOCALOID5 is better at creating whispery or breathy vocals from the default sound. SVS is more like early VOCALOID, where after a certain point breath makes it sound very harsh and buzzy.
No native English "power" or rock-oriented vocals yet from Dreamtonics, they're all kind of soft characters. Here's hoping SOLARIA fills that power niche. And if AI voices are just gonna be soft because that's what the training produces, then a VOCALOID-like "Growl" feature would help a lot to add grit.

ZincT · Post by **ZincT** » Tue Dec 07, 2021 11:50 am

Sounds impressive!

If I want to buy Eleanor Forte AI (English), ANRI AI (English) and Synth V Pro which is the best place to buy and which package do I need? There seem to be multiple links for the same thing (or at least that's what it looks like).

Thanks

Synthesizer V - Singing voice synth for Win/Linux/Mac (Free & Payware)