Emvoice One New Pack and Sale

[Larry Shelby]

Cloud-based singing synthesizer Emvoice One has slashed the base price of their voices and put them on sale with the release of "Thomas", their new Daft Punk-esque vocal.

Mac/PC, VST/AU/AAX, free demo. Requires active internet connection to use.

Emvoice One "Lucy" (English female) = $59, reg. $99
Emvoice One "Jay" (English male) = $59, reg. $99
Emvoice One "Thomas" (Vocoded vocal) = $79, reg. $99

https://emvoiceapp.com/

Lucy and Jay were originally priced at $199.

[ZincT]

Great prices!

Maybe in response to Synthesizer V improvements.

This SV demo uses five different voice banks (mostly Chinese but singing in English) to "simulate" the Spice Girls vocals.

https://www.youtube.com/watch?v=TotFdChJ7i0

 

[Reid Rosefelt]

I'm intrigued by Emvoice, but I do worry about the price drops.  If the company goes under then the software no longer works.

The downside of Emvoice is that you only license voices.  The program is in the cloud.  The upside is that there are free updates forever.  The only thing you pay for would be new voices. 

Black Friday 2020  $139 per voice

Black Friday 2021 $119

Earlier this year $99

Now $59

Even if it goes lower, $59 seems like a very fair price. For the BF 2020 price for one voice you can now get two voices.     if it's as good as the demos.  I'm going to download the demo, and decide for myself, although I read somewhere that it wasn't available anymore.

Hopefully $59 will be the sweet spot and a lot of people may jump.  I have a video planned where I could really use this, so I am likely to be one of them.

 

 

[antler]

I love vocal libraries, and I like the idea behind this. There's something about Emvoice that just sounds artificial to me though: probably because the generated vocal's pitch is extremely precise and unwavering - I guess it's the same as a modern day hyper auto-tuned vocal though.

[PavlovsCat]

Hmm. the Jay voice sounds pretty good and I imagine would probably work to do harmonies with me. Do any of you who own this comment on its ease of use? I heard the demo for Last Town a couple of years ago and always thought about buying this. If anyone remembers an artist who had a hit in the 90s called "Barely Breathing" named Duncan Sheik, this sounds a bit like him and I really like that. Considering that I was a musician who did background vocals that really doesn't have a lead singer quality voice, I was thinking if I could use this for harmonies it could really improve things. But if it's really difficult to use, I doubt I'm going to bother going through the effort.  

EDIT: I am going to download the demo. But I'd still greatly appreciate feedback from someone who's spent time using this. 
 

 

Edited January 24, 2022 by PavlovsCat

[locrian]

This update is a *huge* disappointment!

The three main problems with this platform are, as others have noted, 1) the need to go online to render voices, 2) voice realism, and 3) usability (my biggest gripe).  But instead of addressing these issues they wasted time introducing a vocoded-voice which can easily be created with many widely available, lost-cost, 3rd-party FX plugins.  Why?!?

Meanwhile the editor is still absolutely primitive compared to that of Synthesizer V.  Emvoice lacks such basic functionality as full-screen zooming and note identification so you're constantly zooming in/out when editing.  I find it's easier to just edit the MIDI data in my DAW and then read it back into Emvoice.

They should have spent time adding features like the ability to read in text-files containing your lyrics and basic editor functionality such as scaling note-length independently of neighboring notes.  As it stands now, if two notes are connected, you cannot break them apart and shorten one; you can only adjust the shared onset/offset time.  So instead you have to delete one of the notes, add a new above/below the other note, make the new note short enough to prevent it from getting reconnected to the existing one, and then drag it down/up to the correct note value (which involves counting notes [C#, D, D#, E,...] while dragging the new note b/c the paino roll keys are not labeled).  Needless to say I don't bother with all this hassle so I just do that editing in my DAW as well.  And so you're constantly jumping back-and-forth between Emvoice and your DAW to get things right.  It's a *real* drag.

The plugin also periodically crashes my DAW (Cubase 11 / Windows 10) which makes the experience even more frustrating, time-consuming, and inspiration-killing.

And don't even get me started on the timbre of Jay's and Lucy's voices; neither is usable in most genres.  Even at $59 I will *not* buy another voice.  I've lost faith in this developer.

Time to switch to Synthesizer V...

[Reid Rosefelt]

The Thomas launch sale ends on February 14th.  After that, all three will return to $99.   I bought Lucy and will buy Jay now.  I was thinking about it, but this gets me off the fence. 

I'm working on  a video on Emvoice now.  It includes a section on Emvoice vs Synthesizer V to help people decide which is best for them.  I thought my video would be out by today, but now it looks more like Monday.  Cubase and Voicemeeter!  Arggggh!  ?

I agree with much of what @locrian said.  But you can break notes apart by holding down the alt key.  Maybe he means two notes connected in a legato, where one word is two notes.    He's true about that.   But if you have a group of notes you can select any you want, hold down the alt key, and drag them away.

but Emvoice is still best for me personally, for reasons I'll get into in my video.  For one thing, I have zero interest in pushing the envelope on what a virtual voice can do.  But if that's what you want, Synthesizer V is the future.  But it is also much more expensive today with two voices for $120.  It may be $150 for the Synthesizer V program and a voice, but if you want to get Eleanor Forte, Solaria, etc., then you are looking closer to $200.  

I had very specific ideas for how I would use Emvoice when I got it (similar to my "Betty" on my Betty Page video) but since I got it, it's become an inspiring tool for songwriting.  But if I want a voice, I'll use mine or get somebody else. 

The Synth V free demo is amazing, but it comes with LE versions of voices--so you can't evaluate quality from them.   Emvoice is the same voices as the pro ones, but only 8 notes.

But anybody who is curious can download the excellent free demos of both of the instruments and see for themselves.   Or watch my video.  ?

[locrian]

13 minutes ago, Reid Rosefelt said:

I agree with much of what @locrian said.  But you can break notes apart by holding down the alt key.  Maybe he means two notes connected in a legato, where one word is two notes.    He's true about that.

Hi Reid,

My gripe is that we cannot break apart "...two notes connected in a legato, where one word is two notes."

Currently the only way (I know of) to accomplish this is to delete one note, insert a new shorter one above/below the existing 'block' of notes, and then drag the new one to where you want it.  The cause of this problem is that Emvoice assumes you want to keep the legato transition between notes and thus only allows you to drag-move the boundary between them rather then separate them.

I'll watch your video when it comes out because I find working with Emvoice very frustrating (though I have used it quite a bit!).

Cheers...

[Reid Rosefelt]

On 2/10/2022 at 4:23 PM, locrian said:

Hi Reid,

My gripe is that we cannot break apart "...two notes connected in a legato, where one word is two notes."

 

Totally true.  The interface is not intuitive in a lot of ways.  In most ways, I would say. 

On 2/10/2022 at 4:23 PM, locrian said:

I'll watch your video when it comes out because I find working with Emvoice very frustrating (though I have used it quite a bit!).

In my video I will strongly advise people to create the melody for the first pass of their entire song in their DAW.  I work as I always do, which is to play live and then edit.  But people who can't play can just draw notes, or use whatever tools they have to create melodies. 

After that, there are a lot of things you can only do within Emvoice, like vibrato.   But even with the second stage of editing,  I go back and forth between the Emvoice editor and Cubase.   If it's easier to make corrections and changes in Cubase, I do that.  I use the Emvoice editor only when it's necessary or easier.  

Despite that, I believe that Emvoice is still simpler.  There isn't that much to learn because it doesn't do much. It is also currently a lot cheaper. For me,  Emvoice is the right choice, but I can understand how for many--if not most--the right choice is Synthesizer V.   Even though I only have the Synth V demo,  I try to lay out the plusses and minuses of both.  It's great that I bought my copy of the Lucy voice and don't have to worry about the influence of the  NFR.  I'll say what I want.

Back to editing!

Edited February 12, 2022 by Reid Rosefelt

[msmcleod]

The nicest interface I've seen for "singing" vsti's is Plogue Alter-Ego  (the same guys that do Sforzando), and like Sforzando, it's free.

You type the lyrics in, then it sings a syllable for each separate note you play.  If you play legato, it sings the current syllable over the whole legato phrase.  Occasionally you have to tweak the spelling to get the right word you want, but for the most part it works really well.

The voices themselves sound nothing like a real person, but it's fine for mocking songs up, checking if melodies work, or passing on as a demo to a real singer etc.

[ZincT]

On 2/10/2022 at 8:57 PM, Reid Rosefelt said:

The Synth V free demo is amazing, but it comes with LE versions of voices--so you can't evaluate quality from them.   Emvoice is the same voices as the pro ones, but only 8 notes.

But anybody who is curious can download the excellent free demos of both of the instruments and see for themselves.   Or watch my video.  ?

Hi Reid,

I think your analysis is fair, but I just thought I would add something as I have now jumped in and bought Synth V Pro along with Solaria and Ryo voicebanks.
I guess I would fall into the category of someone who would like to "push the envelope on what a virtual voice can do".  My own voice is just okay, and is not likely to improve as I get older. So having male/female demo singers available to sing my songs as realistically as possible with the least effort seems the way to go for me.

After buying SV Pro and the voicebanks I thought it might be interesting to simply replace Eleanor Forte AI Lite with Solaria in one of my early SV projects, to see if I could tell much difference. Let's just say that I'm still trying to pick my jaw off the floor  I have been following the improvements in SV and AI voicebanks for months, but I honestly didn't expect such a difference between the lite and Pro versions. Maybe some of the differences are to do with them being different voicebanks idk. 

Something I have discovered is the Auto Pitch Tuning feature which I had initially ignored thinking it was like Autotune/Melodyne. What it seems to do is to make the rise to the start of notes and the drop at the end of notes sound more realistic like a human would sing. Some of the results sound so realistic and, even where the notes are actually slightly out of tune, that in itself makes it sound more realistic.

Anyway, I will look forward to watching your video.   

[Reid Rosefelt]

10 hours ago, ZincT said:

Hi Reid,

I think your analysis is fair, but I just thought I would add something as I have now jumped in and bought Synth V Pro along with Solaria and Ryo voicebanks.
I guess I would fall into the category of someone who would like to "push the envelope on what a virtual voice can do".  My own voice is just okay, and is not likely to improve as I get older. So having male/female demo singers available to sing my songs as realistically as possible with the least effort seems the way to go for me.

After buying SV Pro and the voicebanks I thought it might be interesting to simply replace Eleanor Forte AI Lite with Solaria in one of my early SV projects, to see if I could tell much difference. Let's just say that I'm still trying to pick my jaw off the floor  I have been following the improvements in SV and AI voicebanks for months, but I honestly didn't expect such a difference between the lite and Pro versions. Maybe some of the differences are to do with them being different voicebanks idk. 

Something I have discovered is the Auto Pitch Tuning feature which I had initially ignored thinking it was like Autotune/Melodyne. What it seems to do is to make the rise to the start of notes and the drop at the end of notes sound more realistic like a human would sing. Some of the results sound so realistic and, even where the notes are actually slightly out of tune, that in itself makes it sound more realistic.

Anyway, I will look forward to watching your video.   

Sounds great.  Let me know when you post the results of your work with Solaris.  I think we will see a lot more Synthesizer V voices coming out this year and it will be a game-changer.   Who knows?  I may even get it myself.  Working with Emvoice has piqued my interest.  In some ways, despite its advanced feature set, Synth V is actually easier to use than Emvoice.  ?

I'm working through the Super Bowl trying to get this video done and uploaded tomorrow.  It's really not directed to you, because you already know how to use the program.  On the other hand, some of the things I talk about would be useful to anybody using a virtual vocalist. 

[ZincT]

13 hours ago, Reid Rosefelt said:

Sounds great.  Let me know when you post the results of your work with Solaris.  I think we will see a lot more Synthesizer V voices coming out this year and it will be a game-changer.   Who knows?  I may even get it myself.  Working with Emvoice has piqued my interest.  In some ways, despite its advanced feature set, Synth V is actually easier to use than Emvoice.  ?

 

Here's the initial test I did. It's just the first verse and chorus of True Colors (based on the Cyndi Lauper version). 

This is the Eleanor AI Lite and Synth V Basic version --> https://soundclick.com/r/s8jv82

This is the Solaria and Synth V Pro version --> https://soundclick.com/r/s8ker8 

These are mixed in CbB with my usual treatment on vocal tracks (Fabfilter compression, eq and some de-essing). I use the CA-2A on the vocal bus and a send for reverb. When you look at the waveforms in CbB for each vocal track the thing that stands out is that Solaria is much more dynamic.

I can definitely see me buying some more AI voicebanks and even Eleanor Forte AI seems to bring a different tone to the table. 
Here's a song by one of the BIAB forum regulars on SoundCloud which features both Eleanor Forte AI and Solaria. 

https://soundcloud.com/dcuny/the-gates-of-babylon

 

13 hours ago, Reid Rosefelt said:

I'm working through the Super Bowl trying to get this video done and uploaded tomorrow.  It's really not directed to you, because you already know how to use the program.  On the other hand, some of the things I talk about would be useful to anybody using a virtual vocalist. 

I will probably watch it anyway as your videos are always worth a watch  ???

Edited February 14, 2022 by ZincT

[Reid Rosefelt]

lt's currently over 50 minutes long.  It's a complete video manual of Emvoice, plus all my opinions.   I had it up on YouTube and was working on the chapter headings when I realized some vital stuff was missing.  So I'm doing some new shooting today.   It's a hassle, but the video will be a lot better. 

Should be up tomorrow.  Too late for the sale. 

[Reid Rosefelt]

My video is finally up!   It's 57 minutes long, so I don't expect anybody will watch it all, but I've spent a lot of time on the Chapter Markers so people can just pick what they want to see.  One Chapter is a brief comparison of Emvoice One and Synthesizer V,  and there's a lot of stuff that I think will be relevant to anybody working with a virtual vocalist--like using your DAW (including any MIDI utilities you own) to create your melodies, putting your backing track in first, thoughts on legato and vibrato, and so on.

 

[locrian]

3 hours ago, Reid Rosefelt said:

My video is finally up!   It's 57 minutes long, so I don't expect anybody will watch it all, but I've spent a lot of time on the Chapter Markers so people can just pick what they want to see.  One Chapter is a brief comparison of Emvoice One and Synthesizer V,  and there's a lot of stuff that I think will be relevant to anybody working with a virtual vocalist--like using your DAW (including any MIDI utilities you own) to create your melodies, putting your backing track in first, thoughts on legato and vibrato, and so on.

Gee, 1:03 into the video and I already learned something: I never noticed the play button that allows you to preview your vocal.  I've been auditioning by soloing the Emvoice track and playing my entire project.

Apparently I have a lot to learn ... off to school now.

Thanks, Reid

[locrian]

4 hours ago, Reid Rosefelt said:

My video is finally up!   It's 57 minutes long, so I don't expect anybody will watch it all, but I've spent a lot of time on the Chapter Markers so people can just pick what they want to see.  One Chapter is a brief comparison of Emvoice One and Synthesizer V,  and there's a lot of stuff that I think will be relevant to anybody working with a virtual vocalist--like using your DAW (including any MIDI utilities you own) to create your melodies, putting your backing track in first, thoughts on legato and vibrato, and so on.

I finished watching the video and learned a few new tricks that will certainly make using this plugin much easier.

Particularly useful is the ability to:

•  export songs and the user dictionary
•  incorporate specific phonemes (via chevrons)
• save alternative pronunciations

Thanks again, Reid!

[Reid Rosefelt]

1 hour ago, locrian said:

I finished watching the video and learned a few new tricks that will certainly make using this plugin much easier.

Particularly useful is the ability to:

•  export songs and the user dictionary
•  incorporate specific phonemes (via chevrons)
• save alternative pronunciations

Thanks again, Reid!

Thank you!  This makes me feel much better about all the time I spent on this. 

I already found out that I made a big mistake about Synthesizer V.   It seems that Eleanor Forte was released back in 2018, so Synth V has had English-language voices from the beginning.  What's new is that there are a lot of other voices coming out.  Oh well, I do say that I didn't know much about vocal synthesis until very recently. 

[ZincT]

@Reid Rosefelt Great work Reid! You must have put in a lot of work for this. 

Lots of useful tips and ideas to improve workflow. A must watch for Emvoice owners especially but also for anyone considering a vocal synthesis library purchase.  

I think the AI version of Eleanor Forte is more recent (Dec 21) than the non-AI (2018). I have both and, as expected, the AI Lite version is a significant improvement over the non-AI Lite version.

Edited February 16, 2022 by ZincT

[locrian]

They just added a Split Tool...