Synthesizer V Studio: Second Generation synthesized vocals from Dreamtonics
5,0 / 5,0 | 
Dreamtonics Synthesizer V Studio · Source: Dreamtonics
Dreamtonics Synthesizer V Studio · Source: Dreamtonics
Step into the eerily realistic world of synthesized vocals in version 2 of Dreamtonic’s Synthesizer V Studio with realtime rendering and a range of Anime styled vocalists to choose from.
Synthesizer V Studio
“V” is for vocals and Dreamtonics believe they can provide you with a vocalist that’s not going to be late for any sessions, get too drunk to sing or run off with your boyfriend. It’s less of a synthesizer in that you don’t “play it” and it’s more of studio environment where you build vocal tracks. Think of it as Melodyne but in reverse.
The voices are generated by a combination of artificial intelligence and samples which allows it to find a natural and organic tone while giving you the ability to control the voice and the performance. In this new version of the engine it brings in neural networks to enhance the quality of elements such as breaths and whispers.
You need a voice to start with and currently, there are 10 to choose from. They are displayed as cute anime characters, predominately female and described in slightly odd ways. For instance, Chiyu is “A brave and honest girl. Her bright flaming voice reaches far and wide like a bonfire.” and Shian has “The faint smile on her is reminiscent of a baby or a teddy bear, rather. A soft and warm voice is released from the hold of her mouth.” Six of them sing in Chinese, three in Japanese and one in English called Eleanor Forte who “Despite being an epitome of the latest technology, Eleanor keeps an interest in antiques. She is honest, endeavoring, with an unwavering sense of morality.” which is good to know.
So what do they sound like? Pretty cool actually. Of course, there’s a lot of Japanese and Chinese pop music in this and so vocals overladen with autotune effects are not uncommon. But there’s also more to it than that and a surprising level of realism.
Here’s an example which, although synthetic, doesn’t sound like it couldn’t be real:
You are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
This one is a more realistic example in Chinese.
You are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
It’s really quite interesting.
Interface
Synthesizer V Studio itself looks very familiar with a piano roll interface and vocal objects that look every bit like Melodyne. You place notes and the vocals render themselves. Add in lyrics and it automatically throws them onto the notes you’ve placed. You can download a free “Lite” version and try it out for yourself, it’s really quite fun and slightly amazing.
You can tune things, add colour and vibrato, slide from note to note add accents and control breathiness, tone and pronunciations. You can even script up processes to automate your workflow. You can use it standalone or as a plugin within your DAW.
I wonder whether it’s worth all the trouble to craft a vocal performance in software when perhaps you could get your mate to sing it instead. It definitely has possibilities although I’m not sure we’re ready for it on this side of the planet yet.
Synthesizer V Studio is available now for around £80 but try out the free version first to get a flavour of the possibilities. And there’s a long video on the website going into all the details about the technology.
More information
- Dreamtonics website.
This might also interest you
One response to “Synthesizer V Studio: Second Generation synthesized vocals from Dreamtonics”
No reference to Vocaloid?
But this has some impressive results, from what I hear on here. This could definitely be helpful to solo producers. Create some demos until you can find a vocalist for your songs