×
What's new
By:

ACE Studio discussion thread

M

milford59

Senior Member
There seems to be some support for a separate ACE Studio discussion thread so I will start it off with this post by ThomasS - sorry but I haven’t been able to copy the links to the voices - perhaps Thomas could reply and do that , to get the thread off and running.

Thomas Said:


Demo of Three New Native English AI Voices

Ace Studio has a ton of voices (over 30) but only three are native English (David, Lien and Bianca.) Their Asian voices have great tone but can't pronounce English very well. However, using the 14-day trial, I tried out their three English voices:


Lien - Kiss and Say Goodbye


David - Kiss and Say Goodbye


Bianca - Kiss and Say Goodbye

Which, if any, do you like?

Don't judge the pronunciation too much, because I've only had the program for a little over an hour, and didn't tweak the words much. Ace Studio has excellent control over pitch curves, and a better vibrato control (note by note) which can shape the modulation and/or the the frequency as well as the onset, so it is great for this kind of soul music. This took less than an hour, because I let it read and notate the isolated vocal from the original recording, and it copied the pitch curves as well as the lyrics. It didn't get all the words right (kind of like Siri or Google) but it's easy enough to correct if you want. I only did that for a few words, since this was just a test. It fairly accurately copied the pitch curves from the recording (kind of like Praat/Hatori) but some of the vibratos and slides are not steady, so we would need to tweak them here and there. Still, it's pretty good, and you can sing your own vocals to get a human performance very quickly.

I'm going to test transferring files to and from Ace Studio and Synth V using this convertor, to see if we can get these voices triggered by Synth V, which will increase the amount of English voices we have. I don't know if anyone has tested this yet, but if it works it will be great
Click to read more...
 
ip20 said:
Based on Dr. Mix’s video, the audio vocal he recorded to midi convert tool is pretty useful.

Does SynthV have something like that?
Click to expand...
Yes it does have AUDIO to MIDI converter .. includes the lyrics interpretation depends on how clear the pronunciation is ..
 
The Doctor Mix video was/is quite an impressive overview of ACE. The extra "behind the scenes" video on Doctor Mix's website is really cool/interesting too.

I'm trying to decide between ACE and SynthV right now... this thread will be helpful. Thanks all!
 
FrankLP said:
The Doctor Mix video was/is quite an impressive overview of ACE. The extra "behind the scenes" video on Doctor Mix's website is really cool/interesting too.

I'm trying to decide between ACE and SynthV right now... this thread will be helpful. Thanks all!
Click to expand...
ACE has to be online when you are using it (the beta was like that if I'm not mistaken) .. but I think they've grown really fast now. I like ACE but the subscription model made me choose the synth V instead. Synth V also had grown up .. even not as quick as ACE but I think it really has potential.
 
FrankLP said:
The Doctor Mix video was/is quite an impressive overview of ACE. The extra "behind the scenes" video on Doctor Mix's website is really cool/interesting too.

I'm trying to decide between ACE and SynthV right now... this thread will be helpful. Thanks all!
Click to expand...
Is there anything holding you back from making a decision, or does your posted song mean you've made that decision? I like your tune by the way. The backing reminds me a bit of a Queen song... maybe Radio Gaga?
 
richiebee said:
Is there anything holding you back from making a decision, or does your posted song mean you've made that decision? I like your tune by the way. The backing reminds me a bit of a Queen song... maybe Radio Gaga?
Click to expand...
Yeah I’m going with ACE for now. I like the UI and the number of voices I can work with and/or combine to create “custom” voices.

Like any new tech, I’ll keep an eye on the market and see what transpires over the next year.

I must say though, for a writer like me that can’t sing and have limited to zero funds to hire vocalists, this is really a game changer.


And thanks for the compliments! I appreciate you taking the time to listen and offering your perspective!

By the way, the backing vocals are Eastwest Choir using their Word Builder.
 
FrankLP said:
Yeah I’m going with ACE for now. I like the UI and the number of voices I can work with and/or combine to create “custom” voices.

Like any new tech, I’ll keep an eye on the market and see what transpires over the next year.

I must say though, for a writer like me that can’t sing and have limited to zero funds to hire vocalists, this is really a game changer.


And thanks for the compliments! I appreciate you taking the time to listen and offering your perspective!

By the way, the backing vocals are Eastwest Choir using their Word Builder.
Click to expand...
As a non-singer myself, I totally get you. Synth-V basically changed what I do with all of my spare time. :)
 
Hi, whenever I try to use the convert audio to midi tool (Acapella Convert) in ACE I keep getting an error message saying Converting Failed... please retry.
I've tried everything, unsure if I'm doing anything wrong. I am using the trial version for now, maybe it's because of that? Apart from this the program runs fine on my MacBook M2
 
Hi, despite my initial critic (in another ACE Studio thread), I decided to give it a go with a 14-day trial.
My main problem is that my Sibelius Ultimate notation program generated MIDI file with melody and lyrics somehow gets misaligned after imported in ACE. With Synth V I have no problem at all.
I also notice that the pronounciation in ACE is not very clear. Could depend on the voice I chose.
In Synth V this is not the case.

Any tips/tricks to get things right in ACE?
 
It probably gets a bit lost in the synth-v traffic. There’s a thread in the introductions forum started a week or so ago by the developer of Ace. That might be a better bet since he hasn’t ventured to this part of the forum yet.
 
This runs
Phryq said:
This runs on a cloud or locally using GPU?

No VAT I guess?
Click to expand...
This runs in the cloud, but as of version 2.0 there is turbo mode for voice generation, which does run locally if your system can handle it. It does not Apple silicon.
 
You must log in or register to reply here.
Back
Top Bottom