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

[Pelicanomicon]

It's because the tuner is aman0_kei, one of the developers of SV Genbu Standard.

Would you mind explaining to a tyro what the "tuner's" position is in the process? Is he/she a developer or is this "tuning" user accessible in any way?

Ah sorry. The term "tuning" I believe comes out of the VOCALOID userbase: it's the manipulation of all of a vocal synth's parameters to create a vocal performance. So a "tuner" is simply any user who has crafted/sequenced/programmed the vocal track used in a song. There are some musicians who have become famous in vocal synth circles just for their amazing tuning skills, like cillia/kyaami
https://www.youtube.com/watch?v=Wae9v1rxZzA

And Mitchie M
https://www.youtube.com/watch?v=uwwU55zBYlQ

What is the "Genbu Standard", please?

Genbu Standard is a voicebank for Synthesizer V Studio, a Japanese male. "Standard" in SV terms means a non-AI voicebank, they're more like VOCALOID voicebanks.

[Spitfire31]

Thank you for spreading the light, Pelicanomicon!

[Spitfire31]

Kyaami's treatment of Sachiko is phenomenal!

Sachiko is a Vocaloid voice, I gather. I suppose it's no use hoping that she will be an SV AI voice one day? Just amazing…

[Pelicanomicon]

Not surprised that Emvoice have slashed their prices!

Regarding what happens to customers if Emvoice goes under:
Years ago in their own KVR thread, the dev did comment they'd release an offline version but who knows if that is still their plan.

For Thomas's release, Emvoice has been uploading snippets of their improvements to YouTube over the past few hours. They show things like "Pitch Contour Presets", "True Formant Shifting", "Alternate Takes", a new engine and so on. No in-depth details yet on the website.
https://www.youtube.com/watch?v=PhTI3WOm_4U
To my ears, Lucy's pronunciation is a tad more natural now but I think the BIAB comparison still applies.

[Pelicanomicon]

Kyaami's treatment of Sachiko is phenomenal!

Sachiko is a Vocaloid voice, I gather. I suppose it's no use hoping that she will be an SV AI voice one day? Just amazing…

Man, I wish! VOCALOID4 Sachiko is voiced by a famous enka singer, Kobayashi Sachiko. Her live performances are Crazy, with a capital Cuh
https://www.youtube.com/watch?v=l3Y5A-JZvig
I would be so happy if Dreamtonics courted a seasoned enka performer for SV AI.

[ZincT]

Not surprised that Emvoice have slashed their prices!

Regarding what happens to customers if Emvoice goes under:
Years ago in their own KVR thread, the dev did comment they'd release an offline version but who knows if that is still their plan.

For Thomas's release, Emvoice has been uploading snippets of their improvements to YouTube over the past few hours. They show things like "Pitch Contour Presets", "True Formant Shifting", "Alternate Takes", a new engine and so on. No in-depth details yet on the website.
https://www.youtube.com/watch?v=PhTI3WOm_4U
To my ears, Lucy's pronunciation is a tad more natural now but I think the BIAB comparison still applies.

Very interesting; let's see what they come up with. I can see why they would want to improve Lucy and co but I don't understand why they would introduce a Vocoded voice. Aren't we looking for ultimate realism? The inability to distinguish between synthesised and real. At least that's what I'm looking for and Solaria is getting pretty close! But vocoded vocals would seem to be the opposite of that and something that most folk here could probably do without buying a voicebank.

[Pelicanomicon]

If Cross-Lang hadn't happened I wouldn't bother to post this next news, but now that all AI voices are "English" voices...

AHS is flexing again by releasing Kyomachi Seika's Standard and AI Lite vbs a week ahead of commercial launch.

Ruh roh, AHS has apparently released Kyomachi Seika without Cross-Language support People who bought the AI version are reporting that only Japanese synthesis is available.

https://forum.synthesizerv.com/t/topic/4557/36

https://twitter.com/spunky_dad/status/1 ... 6902797313

I had assumed as well that all new AI voices would have Cross-Lang enabled. AFAIK AHS hasn't said anything, so who knows if support is coming in a patch.

[Pelicanomicon]

Crisis averted, AHS has patched in multilingual support for Kyomachi Seika AI!
https://twitter.com/ahsoft/status/1486699717207937026

Here's a brief demo of Dreamtonics' unnamed English male from the latest AHS livestream, timestamp 1:56:44. At 1:58:15, AHS also demos a brand new voice in development: their first Japanese male for Synthesizer V! They'll show a bit more in the next livestream:
https://www.youtube.com/watch?v=i_II7NSZ51Q&t=7005s
Finally, here's Quadimension's Stardust Infinity English AI demo "7rings" (Ariana Grande), tuning by 瑞安Ryan. Stardust was originally released for VOCALOID4 and is voiced by viral singer Chalili.
https://www.bilibili.com/video/BV1i34y117b5

[Pelicanomicon]

It's rainin' MEN!

Today, Dreamtonics announced the release of "Kevin", the first native English male AI vocal for Synthesizer V Studio!
https://www.youtube.com/watch?v=3R4TKDV2Jf0

Synthesizer V AI Kevin (native language: English) is a singing voice database with a wide range, clear high frequencies, and a powerful voice. He is best suited for powerful tunes such as rock and EDM.

...and "Ryo", the first native Japanese AI male vocal!
https://www.youtube.com/watch?v=I6_bY8Cnzvk

Synthesizer V AI Ryo (native language: Japanese) is a singing voice database featuring a warm and gentle voice. He is good at the middle range, ideal for pop music yet flexible enough to be used in various genres.

Official tweet, with Cross-Lingual Synthesis support confirmed:
https://twitter.com/dreamtonics_en/stat ... 2337809408

Also finally available outside China is "Qing Su", a native Chinese female AI voice!
https://www.youtube.com/watch?v=weXbbXKHUDI
Get them at Dreamtonics' brand new store for in-house products! Which got briefly hugged to death, but it's okay now.
https://twitter.com/dreamtonics_en/stat ... 3858046976

Direct link: https://store.dreamtonics.com/

All Dreamtonics voices = $79.00 + tax.
Synthesizer V Studio Pro (the editor) = $89.00 + tax.
[Bundle] Synthesizer V Studio Pro + Any Voice Database = sale $151.20 + tax, reg. $168.00

[ZincT]

Great news! Thanks @Pelicanomicon.

Kevin and Ryo both sound excellent, but I'm swaying more to Ryo (in English thanks to cross-lingual support).

Qing Su also sounds sweet! I guess we are going to be spoilt for choice soon

I have held off going for SV pro until now and the opening of the Dreamtonics store seems like the perfect opportunity to buy a bundle.

[Pelicanomicon]

Crusher's tuning leans toward the uncanny, in Kevin's case it is kinda hard to tell if the demo reflects her style or his default sound
https://www.youtube.com/watch?v=DtKXThAkQnk
I prefer Ryo as well, but it could be because I'm a sucker for that "Tsubasa" song AHS always uses

[Musical Gym]

How do I install the free voice into Basic?

[Pelicanomicon]

How do I install the free voice into Basic?

Usually double-clicking the .svpk file or dragging it into the Synth V Studio window will start the installation dialogue.

From the Synthesizer V forum, "How load voice database in Basic 2021"
https://forum.synthesizerv.com/t/topic/3026

Tip if you're using Linux, "Unable to install Voicebanks (On Linux)"
https://forum.synthesizerv.com/t/topic/4805

[Musical Gym]

Thanks much omicon!

Now I need to figure out why I need this when I already have Vocaloid.

[Ou_Tis]

Crusher's tuning leans toward the uncanny, in Kevin's case it is kinda hard to tell if the demo reflects her style or his default sound

More realistic Kevin:

https://www.youtube.com/watch?v=dqYjrtV7Eag

Think I still prefer Ryo so far---would be interesting to hear Kevin in a lower register and Ryo doing English. I've found the cross-lingual pretty convincing and I like Ryo's darker timbre.