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

VST, AU, AAX, etc. plug-in Virtual Instruments discussion
RELATED
PRODUCTS
Pelicanomicon
KVRist
Topic Starter
158 posts since 29 Sep, 2015

Post Sat Aug 18, 2018 10:46 am

Edited to add info as of 29 May 2022:
From Dreamtonics, https://dreamtonics.com/en/synthesizerv/
Compatible with Windows, Mac, and Linux

https://www.youtube.com/watch?v=MndJGQZUetE
VERSIONS AND WHERE TO GET THEM
  • Synthesizer V Studio Basic: free, feature-limited, NOT time-limited; compatible with free and commercial voices
  • Synthesizer V Studio Pro: paid, supports VST/AU, selectable alternate phoneme recordings, cross-lingual synthesis, Vocal Modes (specific timbre settings like "Airy", "Power"), etc.; compatible with free and commercial voices
SVS Pro downloads officially sold at SVS Pro Starter Pack physical bundle sold at Amazon Japan. Depending on Amazon price fluctuations, where you live and the strength of your currency, the total with shipping can be cheaper than buying the editor and a voicebank separately. SVS Basic freeware and free Lite voices in Chinese, Japanese and English can be downloaded at
https://resource.dreamtonics.com/download/

:!: Lite voices can't be used for profit. If you make your works with them public, you must state what voice you used.
ACTIVATION LIMITS AND USAGE LICENSES

Please see this post for information regarding Synthesizer V Studio Pro install machine/activation limits

Direct link to Dreamtonics' usage license for its first party voice databases
https://dreamtonics.com/en/terms

Direct link to Eclipsed Sounds' license information
https://www.eclipsedsounds.com/solaria-license
See also this post from an official member of SOLARIA's team regarding Eclipsed Sounds' SOLARIA license

Direct link to AUDIOLOGIE's license information
https://audiologie.us/pages/about-us
See also this post for info about AUDIOLOGIE ANRI's license

Please see this post for other information regarding usage licenses for the voicebanks
VOICEBANK TYPES/LANGUAGES AND WHERE TO GET THEM

Synth V has two types of voicebanks: Standard (VOCALOID-like concatenative) and AI-trained

Synth V supports three singing languages: Mandarin Chinese, Japanese, and English. However on 23 Dec 2021 Dreamtonics implemented Cross-Lingual Singing Synthesis, a feature which allows any CLSS-enabled AI voicebank to convincingly sing in any supported language, regardless of the voice provider's original language. :!: CLSS is available only in Synthesizer V Studio Pro, and currently only in commercial (not Lite) voicebanks.

Demonstration of Cross-Lingual Singing-Synthesis:
https://www.youtube.com/watch?v=De4cnZkl5pQ
There are 12 English-based vocals available:
  • From Dreamtonics:
    • Kevin AI
  • From VOLOR:
    • Eleanor Forte Lite (free), a beta Standard voice
    • Eleanor Forte AI
    • Eleanor Forte AI Lite (free)
  • From AHS, Tsurumaki Maki (弦巻マキ), voiced by Japanese actress Tanaka Manami:
    • Tsurumaki Maki English
    • Tsurumaki Maki English Lite (free)
    • Tsurumaki Maki English AI
    • Tsurumaki Maki English AI Lite (free)
  • From AUDIOLOGIE:
    • ANRI AI, voiced by kurapichu
    • ANRI AI Lite (free)
  • From Eclipsed Sounds:
The following non-English AI voices are capable of singing English via Cross-Lingual Singing Synthesis:
  • From Dreamtonics:
    • Saki AI (Jpn female), also sold by official partners AniCute and AHS
    • Ryo (Jpn male), also sold by AHS
    • Qing Su (Chn female)
    • Mo Chen (Chn male)
    • An Xiao (Chn male)
    • Feng Yi (Chn female)
  • From AHS:
    • Koharu Rikka AI ("Synthesizer V 小春六花 AI") (Jpn), voiced by actress Aoyama Yoshino
    • Tsurumaki Maki AI Japanese ("Synthesizer V 弦巻マキ AI"), voiced by actress Tanaka Manami
    • Tsuina-Chan AI ("Synthesizer V AI ついなちゃん") (Jpn), voiced by actress Kadowaki Mai
    • Kyomachi Seika AI ("Synthesizer V AI 京町セイカ")(Jpn), voiced by actress Tachibana Rika
    • Natsuki Karin AI ("Synthesizer V AI 夏色花梨") (Jpn), voiced by actress Takagi Miyu
  • From Quadimension:
    • Xingchen Infinity, aka "Stardust" ("星尘 Infinity") (Chn), voiced by viral singer Chalili
      --------------------
      Currently sold only in China through Taobao. Overseas users have reported successfully using the proxy service ParcelUp (formerly TBFocus) to buy digital copies of Quadimension voices.
      --------------------
      Link to ParcelUp's proxy page for Quadimension Taobao as of 27 May 2022:
      https://parcelup.com/shop/item.php?id=675192462436
      --------------------
      Link to Synthesizer V Forum discussion which notes that after buying through ParcelUp, Quadimension directly emails customers their digital goods:
      https://forum.synthesizerv.com/t/topic/5194
      --------------------
      :!: For reasons unknown, the option to buy Stardust sporadically disappears from the ParcelUp page. If it's not there, try again later.
Also in the works for 2022:
  • AHS Hanakuma Chifuyu AI: Jpn female (October 2022)

LEARNING RESOURCES

Dreamtonics official YouTube video tutorial playlist. Tip: Slow playback down, the demonstrator moves the mouse very fast.
https://www.youtube.com/playlist?list=P ... 0SGRJgRIPz
Please see this post by KVR user koukouba for more information regarding the differences between Basic and Pro editions.

See this post for YouTube video tutorials made by other Synth V users.

Link to Synthesizer V Studio phoneme list for Chinese, Japanese and English
https://forum.synthesizerv.com/t/topic/3249
SVS also supports [pau] and [sil] phonemes.

Online manual for the previous generation of Synthesizer V detailed below, references removed features but has some useful info
https://synthesizerv.com/manual/

~~~~~~~~~~ End edit ~~~~~~~~~~

Synthesizer V
Demo available at https://synthesizerv.com/

Vid from 2017:
https://www.youtube.com/watch?v=aW6x6EMACbg
Based on a hybrid of artificial neural networks and concatenative synthesis, Synthesizer V delivers natural voice from a small amount of sampled data. Our patent-pending Low Level Speech Model (LLSM) separately processes vocal folds and vocal tract features, thereby allowing for high-fidelity and flexible manipulation of voice timbre.
Synthesizer V Technical Preview +
Eleanor Forte (English voice) Bundle

This technical preview offers a peek into the next generation synthesizer’s capabilities. On supported platforms, it runs out-of-box without installation. The package also includes a quickstart guide covering the most frequently used features.

Operating System
Microsoft Windows 7 or above (32/64 bit)
Ubuntu 16.04 or above (64 bit)

Disk Space
300 MB
Last edited by Pelicanomicon on Tue May 31, 2022 12:34 am, edited 66 times in total.

Kelvin92
KVRist
81 posts since 13 Dec, 2017

Post Sun Aug 19, 2018 7:22 am

Is this a standalone or also vst?

Pelicanomicon
KVRist
Topic Starter
158 posts since 29 Sep, 2015

Post Sun Aug 19, 2018 9:21 am

Kelvin92 wrote:Is this a standalone or also vst?
Standalone only. It allows for importing an instrumental to time the vocal. What is it about this type of syntesis that makes DAW integration so difficult, I wonder?

Comments after a brief tryout:

this will feel familiar to anyone who's dabbled with older versions of Yamaha's VOCALOID.

I like that it's portable, no installation required. Download is tiny!, only 116MB including one female voicebank. They're using a typical anime mascot to promote it, but the voice isn't egregiously, uhh, "anime-sounding." IMO English pronunciation is on par with VOCALOID4 Cyber Diva, but more...musical? Less "talk-singy", which is :tu:

Really lightweight, it renders edits FAST

Doesn't recognize my ASIO device, only options are Windows Audio and DirectSound.

Synthesis parameters are
  • Pitch deviation
    Vibrato envelope
    Loudness
    Tension: = VOCALOID1 Harmonics and VOCALOID5 Exciter, simulates intensity of the vocal
    Breathiness
    Voicing: = VOCALOID2-4 Openness and VOCALOID5 Mouth, simulates openness of the mouth/throat
    Gender: aka Character in VOCALOID5, a formant shaper
No Growl analogue.

English phonemes are based on Arpabet coding. No documentation in the guide, you are gonna have to wiki it
https://en.wikipedia.org/wiki/ARPABET

Lyric input feels less elegant than Vocaloid, which smartly distributes syllables of a single word over notes of different pitches and assigns the phonemes. So for example if I'm directly entering lyrics over four notes and type "Daisy [tab] Daisy [tab]", Vocaloid will break this up per note as "[dh eI] [z i:] [dh eI] [z i:]" for me. AFAICT, Synth V won't do this? I have to 2xclick the first note, type "Day", and click off; 2x click the second note, type "zee", click off, etc. --ETA: :dog: watched the promo vid again, you can 2x click successive notes to enter but this is still more cumbersome than Vocaloid's tab to next note.

one pleasant advantage is "Sync-Parameter" mode. With this enabled, you can move selected notes AND their associated parameter envelopes together. Still can't do that in Vocaloid.

All in all it's pretty impressive for a preview, you're basically getting a free Cyber Diva. If nothing else, you can always use it as a modulator for vocoding :)

User avatar
wagtunes
KVRAF
20814 posts since 8 Oct, 2014

Post Sun Aug 19, 2018 9:54 am

So Vocaloid is still the Rolls Royce of vocal synthesis.

mevla
KVRAF
2966 posts since 3 Nov, 2015

Post Sun Aug 19, 2018 10:44 am

This sounds great ! The English voices are great, Japanese quite good also. The Chinese male voice not so good, but the Chinese female voice is much better. And it's also on Linux.

Not sure if this is ready to be used, but this is certainly extremely promising ! Please make it in such a way that it can be easily used in a common DAW.

Don't need no Rolls-Royce.

Your TED talk is also interesting.

Cheers.

User avatar
NoirSuede
KVRist
47 posts since 20 Jul, 2016 from Tangerang, Indonesia

Post Sun Aug 19, 2018 12:04 pm

wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.

User avatar
wagtunes
KVRAF
20814 posts since 8 Oct, 2014

Post Sun Aug 19, 2018 12:07 pm

NoirSuede wrote:
wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.
Sonically, I don't think it's as good as Emvoice. Of course given it's not a VST, doesn't matter. Have no desire to even download it.

Someday Vocaloid will be topped or they'll finally get it as right as they can. Either way, that day has not yet come and these two don't even come close.

User avatar
NoirSuede
KVRist
47 posts since 20 Jul, 2016 from Tangerang, Indonesia

Post Sun Aug 19, 2018 12:08 pm

wagtunes wrote:
NoirSuede wrote:
wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.
Sonically, I don't think it's as good as Emvoice. Of course given it's not a VST, doesn't matter. Have no desire to even download it.

Someday Vocaloid will be topped or they'll finally get it as right as they can. Either way, that day has not yet come and these two don't even come close.
Have you tried CeVIO yet? they just added English localization.

https://www.youtube.com/watch?v=C1GL3LFYJas

User avatar
NoirSuede
KVRist
47 posts since 20 Jul, 2016 from Tangerang, Indonesia

Post Sun Aug 19, 2018 12:36 pm

No VST is pretty painful for us producers, but you guys gotta remember that SynthV is built and targeted for the casual Vocaloid users whom only makes covers and doesn't even use a DAW most of the time, so no VST is a non-factor to them.

User avatar
el-bo (formerly ebow)
KVRAF
16308 posts since 24 May, 2009 from A galaxy, far far away

Post Sun Aug 19, 2018 12:53 pm

NoirSuede wrote:
wagtunes wrote:
NoirSuede wrote:
wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.
Sonically, I don't think it's as good as Emvoice. Of course given it's not a VST, doesn't matter. Have no desire to even download it.

Someday Vocaloid will be topped or they'll finally get it as right as they can. Either way, that day has not yet come and these two don't even come close.
Have you tried CeVIO yet? they just added English localization.

https://www.youtube.com/watch?v=C1GL3LFYJas
Blimey :o :tu:

User avatar
NoirSuede
KVRist
47 posts since 20 Jul, 2016 from Tangerang, Indonesia

Post Sun Aug 19, 2018 1:00 pm

el-bo (formerly ebow) wrote:
NoirSuede wrote:
wagtunes wrote:
NoirSuede wrote:
wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.
Sonically, I don't think it's as good as Emvoice. Of course given it's not a VST, doesn't matter. Have no desire to even download it.

Someday Vocaloid will be topped or they'll finally get it as right as they can. Either way, that day has not yet come and these two don't even come close.
Have you tried CeVIO yet? they just added English localization.

https://www.youtube.com/watch?v=C1GL3LFYJas
Blimey :o :tu:
she's pretty neat, isn't she? she's pretty nostalgic to me bc she's the one that got me into music production haha

User avatar
el-bo (formerly ebow)
KVRAF
16308 posts since 24 May, 2009 from A galaxy, far far away

Post Sun Aug 19, 2018 1:14 pm

NoirSuede wrote:
el-bo (formerly ebow) wrote:
NoirSuede wrote:
wagtunes wrote:
NoirSuede wrote:
wagtunes wrote:So Vocaloid is still the Rolls Royce of vocal synthesis.
so what do you think of this so far?

tbh the most interesting thing so far is how contrasting it is with Emvoice, because while Emvoice is trying to look clean and professional, SynthV is wholly embracing the Vocaloid community that's built up over the years.
Sonically, I don't think it's as good as Emvoice. Of course given it's not a VST, doesn't matter. Have no desire to even download it.

Someday Vocaloid will be topped or they'll finally get it as right as they can. Either way, that day has not yet come and these two don't even come close.
Have you tried CeVIO yet? they just added English localization.

https://www.youtube.com/watch?v=C1GL3LFYJas
Blimey :o :tu:
she's pretty neat, isn't she? she's pretty nostalgic to me bc she's the one that got me into music production haha
Emvoice already seemed to be a clear improvement on Vocaloid, but this seems to push the envelope even further :tu: The chorus is almost flawless.

User avatar
NoirSuede
KVRist
47 posts since 20 Jul, 2016 from Tangerang, Indonesia

Post Sun Aug 19, 2018 1:19 pm

el-bo (formerly ebow) wrote: Emvoice already seemed to be a clear improvement on Vocaloid, but this seems to push the envelope even further :tu: The chorus is almost flawless.
Ironically the engine itself came out wayy before Emvoice, around 2013-ish:
https://cevio.wikia.com/wiki/Sato_Sasara

ENV1
KVRAF
2797 posts since 31 Aug, 2011

Post Sun Aug 19, 2018 1:20 pm

el-bo (formerly ebow) wrote:Emvoice already seemed to be a clear improvement on Vocaloid, but this seems to push the envelope even further :tu: The chorus is almost flawless.
Agreed, thats probably the best synthesized female voice ive heard so far.

Real good.

User avatar
fisherKing
KVRAF
2815 posts since 6 Aug, 2009

Post Sun Aug 19, 2018 1:29 pm

i know, for myself, i would never use any plugin, no matter how good, to replace a real singer. i want it to be another effect, an 'android' voice, or... some combination of those things.

i can see, tho, for song demos, the benefit of having a 'better' voice...
Last edited by fisherKing on Sun Aug 19, 2018 1:30 pm, edited 1 time in total.
https://upstatebrooklyn.com
one plugin short of perfection...

Return to “Instruments”