7 Vsynths that I Really, Really Want Made

Oh, hi! ☕️

Today I'd like to keep things casual. Previously I made a comparison post where I compared VOCALOID to Synthesizer V from a prospective buyer's standpoint. But today, I'm going to be a little more bold.

I'm actually going to share my personal opinions that may or may not be unpopular. 😱️

Shocking, right? Here's my first bombshell opinion: As a program, Synthesizer V is better than VOCALOID.

VOCALOID was an important milestone. It paved the way for Vsynth culture to even exist at all, and for how it exists today. I mean, who knows how different our world would be if Hatsune Miku didn't appear on David Letterman!

You could say that VOCALOID walked so Synthesizer V could run. That Synthesizer V was built on the shoulders of giants. And it was! But I think Synthesizer V is an example of the student surpassing its master. The software is intuitive, reasonably priced, and the AI vocals it can produce are unparalleled.

So why am I telling you this?

Because I want to make it clear that I would really love to see most of these voicebanks made for Synthesizer V specifically. Then all of my dreams would come true! 🥰️

So! Without further ado let's get into it. These are some vocal synths that I would love to see made in the future. But I know realistically, they might not ever come to fruition. But one can dream! 🌠️
 

#1: MIRIAM AI

MIRIAM as depicted on the VOCALOID 1 product box art

Based on the British vocalist Miriam Stockley, MIRIAM for VOCALOID 1 ran out of serial codes a long time ago. But I think that MIRIAM sounds better than a lot of other English Vocaloids that were released later on! In 2017 Miriam Stockley expressed support for an "up-to-date version" of her Vocaloid. But Zero-G Ltd. has indicated in both 2020 and 2023 that they don't have plans to produce any new VOCALOID products. So I think it would be a great idea for MIRIAM to find new life in Synthesizer V with a different production company. I could potentially see Eclipsed Sounds or AUDIOLOGIE making it happen if there was enough interest. Here's hoping! 🤞️

#2 MEIKO & KAITO AI

MEIKO, left, and KAITO, right, as depicted in Project SEKAI

I definitely don't see Crypton going back to third party software, but I can still dream! (Update 8/1/24: I was wrong—they're making Hatsune Miku AI for VOCALOID 6 this year apparently. 😅) I think it would be great to see all of Crypton's Vocaloids on Synthesizer V, but I personally would love to see MEIKO & KAITO especially. Unlike Miku, Rin/Len, and Luka, their voice providers are not voice actors but professional singers. So I think there's a lot of musically expressive potential in AI versions of their voices!

#3: GlaDOS AI

GlaDOS as she appears in Portal 2

The cynical psychopathic computer antagonist from Valve's Portal game series, GlaDOS is truly an enduring character. And she sings! She's voiced by Ellen McLain, a voice actress and opera singer. The ending song from Portal 1, "Still Alive", written by Jonathan Coulton and performed by McLain as GlaDOS, is one of my absolute favorite songs from any video game. And the Italian opera song from Portal 2, "Cara Mia Addio", also performed by McLain as GlaDOS, is really great too. So how cool would it be if you could have GlaDOS sing whatever song you wanted? In English and Italian? It probably won't ever happen, but it's cool to think about Valve software, Ellen McLain, and Eclisped Sounds all collaborating on making an AI voice library for Synthesizer V. That would be so cool... 🥺️

#4 Frank Sinatra-Style AI

Frank Sinatra circa 1957

We got Big Al for Elvis Presley. But I don't know if there's ever been any kind of Vocaloid produced that tries to evoke vibes of a of big band, swing/jazz type singer. This kind of vocal would lend itself really great to Electro Swing music! I feel like it would be really feasible for a future voicebank. Please, someone, make it happen! 🙏️

#5 Choir AI

Stock image, edited

Baba Roy Shuzo has some interesting tracks he's shared from his hard drive from the early days of VOCALOID. One of them is a demonstration of what seems to be a prototype choral VOCALOID 1 library that was never released, singing "Dies Irae" by Mozart. To me this sounds amazing, especially considering this is from the first iteration of VOCALOID. But it was never released! What a tragedy!

While you can definitely find some choral sample libraries to buy, they're very expensive and don't really go beyond some "oohs" and "aahs". I think there would be a lot of potential for having an AI choir voicebank that you could actually input lyrics into. Maybe this would be a good candidate for Dreamtonics to develop themselves?

#6 DECtalk (A.K.A. The Singing Astronauts from Moonbase Alpha)

A blue astronaut from the game

DECtalk Software is one of the oldest speech synthesis programs around, dating back to the 80s. It has the added bonus of being able to sing. The software was implemented into the chat function of a free-to-play moonbase simulation game, Moonbase Alpha, developed in collaboration with NASA. So if you typed words (or singing instructions) into the chatbox, the DECtalk voice in the game would read and/or sing what you typed. This was definitely a recipe for widespread shenanigans and popularity surrounding the game. Many people have made many, many song covers featuring DECtalk, associated with images of the astronauts depicted in the game!

The problem is that it is very difficult to actually make DECtalk voices sing because you have to manually enter timing (in milliseconds) and pitch (with numbers) in plain text for every single syllable (or phonetic sound, if you want to have more control over the timing). Furthermore, any demos of DECtalk Software that you can find online probably aren't officially sanctioned. And if you wanted to use the voices for commercial music production, you'd probably have to try and get in touch with SpeechFX Inc., assuming that you even can (they have a Linkedin page linking to a website that appears to be broken). While Plogue's Chipspeech features a voice that ostensibly is sourced from DECtalk, it sounds too derivative to be immediately recognizable as DECtalk at all. So what do I want exactly?

I'd love it if DECtalk software was somehow integrated into an existing vocal synthesis UI for the purpose of singing. I think it would be cool if the software could be licensed by Dreamtonics and incorporated as an alternate format of vocal library into Synthesizer V. (Having the speech function for rap would be amazing as well). But a standalone DECtalk singing and rap program would also be great. But alas, this is probably just a pipe dream of mine. 😔️

#7 GAGALOID (Lady Gaga VOCALOID)

Lady Gaga, edited image; original by Rogue Artists

Okay, okay. I know this one's the least likely to ever happen on this whole list! But hey, we know that if you're gonna invite a digital pop star to open for you at your concert, you must be a fan. So I wonder if Lady Gaga herself would be open to having a voice library instrument being created based on her voice? With the advent of AI, I'm sure many high-profile commercial singers (or their agents) would be strongly opposed to their voice being used in a singing software that could compete with them for work. But maybe she would be cool with some kind of sample-based library where her voice is altered in some way? Like for VOCALOID 5 or something? Maybe?

Well anyway, here's an old reupload showcasing what a Lady Gaga vocal synth could sound like 😉️


So there you have it! These are some vsynths that I would love to see in the future. Yes, I know some of these have zero chance of seeing the light of day. But it's fun to imagine. 🌈️✨️
 

But what do you think? Do you have any ideas for vsynths you'd like to see made? Let me know! 

That's all I've got for today! Until next time 👋️