It looks like quite a few of us are using this software so maybe we need a thread for it.

I just found this short on YouTube of a voice I had not encountered - Kevin.

Kevin

Yup, that's the one I have. Pretty good male vocal sound.

I talked to David Cuny. Evidently, there are foreign language databases that are useful in English.

I would like to have the AI sing in Spanish.

Here are some lyrics that ChatSonic delivered.

Eh, eh, eh
Una chica guapa con un cuerpo aplaudido
Su cara, su sonrisa, enamora
Su chica linda, su cabello dorado
Alegría, diversión, su carisma de destellos


Baila conmigo, papá, papá
Baila conmigo, papá, papá
Baila, baila, baila, baila


Su chica guapa, la niña de mis ojos
Sus pasos me encienden, imposible escuchar las palabras
Ella me sonríe, me vuelve loco
Su cintura, sus movimientos sensuales


Baila conmigo, papá, papá
Baila conmigo, papá, papá
Baila, baila, baila, baila


Sus giros, sus movimientos, me dejan sin aliento
Sus labios, su perfume, me hipnotizan
Toca mis manos, toca mi cuerpo
Me lleva en volandas con su amor divino


Baila conmigo, papá, papá
Baila conmigo, papá, papá
Baila, baila, baila, baila

Translation to English. I speak Spanish but not well. If any of you can do a better translation please do so.

uh huh huh
A pretty girl with an applauded body
Her face, her smile, falls in love
The pretty girl of hers, her golden hair
Joy, fun, his charisma sparkles


She dances with me, daddy, daddy
Dance with me, daddy, daddy
Dance, dance, dance, dance


The pretty girl of hers, the apple of my eye
Her footsteps turn me on, impossible to hear the words
She smiles at me, she drives me crazy
Her waist, her sensual movements


Dance with me, daddy, daddy
Dance with me, daddy, daddy
Dance, dance, dance, dance


Her twists, her moves, they take my breath away
Her lips, her perfume, hypnotize me
She touches my hands, she touches my body
She carries me on the wings with her divine love


Dance with me, daddy, daddy
Dance with me, daddy, daddy
Dance, dance, dance, dance

Billy

Yes, it's an amazing piece of software and has opened the door for me to write songs with vocals again without depending on unreliable pitchy vocal divas.
My last 10 or so songs have all been done with Synth V.

I own Kevin but am a bit disappointed with it, at least for the style of music I write.

I considered getting Kevin but have not been that impressed by what I have heard so far. I will stick with Solaria, Eleanor Forte AI and Natalie, they sound far more realistic.

I think part of it is definitely how much time you want to spend tweaking each syllable and phrase to fit your music. The more editing, the more realistic. None of the voices sound perfect on MIDI melody import. There's always some work that needs to happen.

Do any of the male English voices allow for more edge/scream as in Rock styles?

Try Dreamtonics NineZero, this sounds very like Ozzy Osbourne to me.

https://www.youtube.com/watch?v=6kzGgaTz3n8

So, what's the difference between Synth V and Synthesizer V? Can the vocal synth run on local computer without constant internet connection?

The Dreamtontics website displays two versions: one older version, one AI version.

My studio's computer doesn't connect to internet.

When you buy the software, can you download the vocal library to your local computer, or you need to connect to internet to access their AI database?

As far as I understand it Synth V & Synthesizer V are one and the same. It's usually referred to Synth V on other sites like Wiki etc.

If you are referring to the two versions for the voices, the standard voice is often known as Lite, sometimes free. The A.I. version is the full version.

You will need to be online to use the voices as they refer back to the Synthesizer V site in real time to access the database.

This is so wrong that it is almost right again

The voice databases are installed locally on your hard drive, and Synthesizer V runs without a network connection. I know this because my DAW is usually offline.
When installing (or updating) the app and/or the voice databases, an internet connection is more convenient, but also not necessary.

My mistake, it is Emvoice that needs to be online to access the database as I recall.

This entire Synth vox thing has got me interested because it has come so far from the early days. I'm hearing some really cool things on Youtube that is simply astounding.

I picked up the basic and one of the lite voice databases just to play with it and see if I want to jump into the deeper end of the pool.

@Planobilly It's an advantage to be able to speak the language well that SynthV singers are supposed to sing. Since SynthV works with phonetic sounds, almost any language (with an accent) is possible, I did it with Asereje and it sounds quite useful. However, it involves a lot of manual rework. You can also simply sing the lyrics yourself and then import them into SynthV using Praat (a free add-on tool).

Is Praat developed by Dreamtonics or a third party developer?

My understanding is, each bot can only sing one language, and you need to type in the words/characters of that particular language in order for the bot to work.

For example, Natalie and Solaria can only sing English. Cheng Xiao and Feng Yi can only sing Chinese. Natsuki and Ryo can only sing Japanese.

Also, their pricing mode is confusing. Each voice database costs $79 to $89. But if you buy the Bundle of Studio Pro + Any Voice database, you can get all voices with only $151?

Studio Pro+ is their advanced editor. I bought that with my first voice for the $151. That price is for the editor plus ONE voice. Any additional voices I've bought have been for the $79-$89.

Praat is an open source software that can be used to analyze almost any audio data.
Synthesizer V users mostly use it when they make covers of songs, you can extract vocal data (phrasing, intonation and whatnot) from the original and apply it to a Synthesizer V singer.
You can also use your own voice (singing that cover) to modulate a virtual singer.
In my opinion, a waste of time for people who write original songs.


Any voice (from the ones available). Not all voices (of the ones available).

@MusicVillain
Praat is free: https://www.fon.hum.uva.nl/praat/download_win.html

Each voice can also sing multiple languages. All words are translated into sound parts and transitions of these parts.
You can download the trial at https://dreamtonics.com/synthesizerv/
receive.

The paid version then has vst, script, AI functions, better voice selection, more tracks, faster rendering and one or the other that I can't think of right now .

Each voice bank is around $80 USD and so is SynthV Pro (comes with Mai Japanese voice bank free).
Natalie, Asterian; Kevin and Solaria were recorded in English, Mai has a small? accent.

Soon the program will also be able to rap. I think the beta will come soon.

great thread! im seriously thinking - planning in fact - to buy this app - although the add-on singers arent cheap...

I am seriously looking at the software as well. My thinking is to use it to possibly do background singer voices to enhance my vocal leads. Don't know if anyone here has tried that yet, but if you have, would appreciate some feedback if creating background singer tracks could work effectively.

Hi there. I have been using Synthesizer V for about a month now and I"m very pleased. I originally bought Solaria and Asterian, since I was cataloguing a lot of music I did for theater years ago.

After doing several songs with Solaria and Asterian, I decided to get Kevin to have a little more vocal variety. However, out of the box, Kevin sounds like a nasally 18 year old, and it seems that his voice works best as a high tenor in pop songs of a certain kind.

However, since I was doing theater stuff, I moved the gender button to the right to get a more mature voice. All of a sudden, Kevin has a european accent. You can't get rid of it. I wrote to Dreamtronics, sent them examples, they claimed there's nothing wrong with the voice. My guess is that they can't hear it.

Has anybody experienced this using Kevin to have a more meaty voice? Anyway, just wondering.

If anybody is just starting out with Synthesizer V, I can assure you that if you have the right voice bank, you can get amazing results, as lead vocals, or background vocals. However, stick with Solaria and Asterian as a start, they truly have English voices. Asterian may not work for all kinds of music, but he has a great voice. Solaria is far more versatile, can work with all kinds of music.

I don't think Kevin is ready for prime time. My opinion from my experience. I think he works if you're doing light pop. I have yet to test out Natalie, I hope I'm not disappointed. Let me know what your experiences with Kevin have been. Thanks.

At least for my genre of music Kevin is/was a waste of time and money.

Same for me, far too wimpy sounding. I'm considering getting NineZero, that one seems very powerful. I would be interested to hear from anyone using NineZero.

He sounds a little tougher than Kevin plus there's a whole bunch of sound files too.....

I need some feedback on this AI vocal and Alt Rock track

Quoted from the post above:

Someone in this thread said if you don't work with English words but with phonic syllables, you can make any bot sing any accent.

Guess that's how you tranform Kevin from an Irish farmer to an American Congresswoman?

Yes, it was for me too. I was just doing another song, just to give this voice another try, but also, emphasizing the gender to make him sound more mature, and in this song, a standard american folk song, the accent is less, but you would never use this guy on a country song.

NineZero is supposedly based on an australian heavy rock singer. I've heard some good productions using his voice. My concern is, is this voice going to have the same problem as Kevin where it only sounds good in some instances. Is this because Dreamtronics is producing the voices, or do you really have to coax a performance out of the voice? Are you going to run into that same problem again.

For instance, Solaria and Asterian are produced by a completely different company. They did it right. With normal editing, those voices can shine, again depending on the type of music you're doing. Right now, I'm doing theater music, and these two sound good. Solaria can sound good on a lot of different types of music though.

So, in answer to your question about NineZero is, I feel that Kevin is useless for the music I'm doing now, is NineZero going to be a repeat of the same? They really need to put out a "lite version" to try before you buy.

The trouble is, for Dreamtronics, that most of the people that use these voices are hobbyists, kids mostly, and if they can use a free voice that sounds half way decent, they're going to stay with using the free voice and never buy the full voice.

I know that they're a small company, but they need to have some kind of refund if you're not happy with the voice. It's $80 bucks for these voices, so if it doesn't work, you're out the money. And the demos they have aren't selling the voice. What sold me was the covers that people are doing on youtube. I realize now that all the productions for Kevin are high tenor pop songs. Kevin ain't never going to do a country song...

Well, there's one guy on youtube that's doing amazing work with different voices and having them sing in Spanish. I've asked him what he's doing he never responds. But he must have a lot of time on his hands and great ears.

The singer goes by the moniker "Ninezero", is based in Japan and speaks Japanese, which probably factored heavily into why he was selected.


Ninezero is even less flexible than Kevin.

You can adjust parameters like Gender that will allow to adjust the position of the formants, but it's still going to sound very much like the singer.


This reflects the skill of the singers, as well as their vocal styles. The problem with targeting a specific style is the voice can't do a lot.

Eclipsed Sounds has put out some very good voicebanks. Being native speakers helps, as does their background in using synthetic voices prior to forming the company.


You've already noted the issues with doing this. It's not cheap to do, and it's up to each company to decide if they want to do that.


Refunds are a pain to process, and there's no way to externally retract a license with the software. They obviously would rather take the hit.


English isn't their primary market. Even Natalie, which purports to be a "native" English speaker, has a strong accent.

The female voices are much better than the males

This is not true. You do not have to be online to use Synthesizer V. Only to download the install program and voices. I believe you are thinking about the Emvoice synth program and yes this one does have to be online to use.