indeed this is a very laudable aim but I’ve no idea how long it will take or whether the results will meet expectations. There was a suggestion a fair amount would already be ready in 2-3 months which strikes me as optimistic but I’d like to be pleasantly surprised.
@MarkSealey, I got an email from Cantai with that as some sort of “motto.” My hope is that Richard is actually aiming for “outstanding” instead of just “good”!
Considering how much they got done between the “beta” and the release, I’m optimistic as well. Lots to sort out and do, however! 
Dear All,
If you don’t want to damage your speakers with Cantai, be careful not to use the “suppress playback on passes” feature, because it currently produces some very unpleasant noises and is not usable at this time.
Has anyone been able to export audio in a project with two flows, different players in each flow? (First flow, reduced choir S/A on one stave, T/B on the other. Second flow, Sop, Ten, Alto, Bass individual staves. Both with a Etude elements piano part). No matter what settings I use in the export dialog I can’t get it to export anything but the piano part.
Same here. A workaround is to export each flow into a single-flow project.
By the way I just got a mail about an update, however the flow-flaw is not being mentioned in the release notes.
V2.2.0 just dropped
Reliability & Activation
- Reinstalling the rendering engine no longer requires you to re-activate the plugin. CantAI now recovers your login automatically when the local server restarts
or is reinstalled. - Your activation is no longer lost if the rendering engine is briefly unreachable. The plugin holds onto your credentials across temporary connection drops.
- Renders now start automatically — you no longer need to open the plugin window after the rendering engine activates in the background.
- Opening a Dorico project with multiple instruments no longer triggers multiple login dialogs. Only one activation prompt appears per session.
- Deactivation from the Settings screen now works correctly.
Windows
- Resolved UI crashes and visual glitches affecting some Windows users.
- Fixed a bug where cache files were being saved to the wrong folder on Windows, which could cause rendering issues.
Rendering
- Improved connection reliability to the rendering engine, reducing cases where the plugin could not reach it on startup.
- Fixed corrupted audio output when multiple render requests arrived at the same time.
- Fixed a token validation error caused by minor clock differences between machines — renders were occasionally failing with a login error despite being correctly
activated. - The “Server starting” banner no longer stays on screen once a render is in progress.
Audio Quality
- Improved pronunciation for contractions and words with apostrophes, such as “I’m,” “I’ll,” “don’t,” and “can’t.”
- Fixed a crash triggered by certain punctuation marks in lyrics.
Voices
- Jon has been moved to the Tenor voice group for more accurate voice matching.
- Bass voices are now labeled “Bass/Baritone” to better reflect their range.
- Aji has been renamed (previously “Aji JP”) and corrected to the Alto voice type.
With a VST pkg on the Mac filesystem which appears to be twice as large as 2.1.0
Same here. Very. It has all the feeling of a successful product.
I was afraid of that. The most recent update does not solve this issue.
This is still rendered wrong…
2.2 has been a mess for me. I am on Windows 10. I installed the 202 file. Windows Defender complained, and I had to force it to install over Defender’s objections. That did not happen at 2.1.
The first time I started Dorico and loaded my score that had a Cantai player, it just locked up solid.
I rebooted and things worked – for about 3 minutes. Then Cantai went silent. The light in the VST panel flashed whenever it was receiving words from Dorico, but silence.
I uninstalled Cantai and reinstalled 2.2 No good – still silent. I looked that the h status in (…). It showed the rendering engine as connected, but cache had zero bytes, and never got anything after that.
So I bailed out and reinstalled 2.1. That is working as before, and had no Defender objections.
Having read this tread and all of the other related threads, I am truly thankful that I decided to wait. If and when all the lice are worked out (I never say bugs), then maybe.
There are still things to fix, yes. trying to export a flow in the middle of a project is blending all the vocal parts of all the flows together In the audio, you don´t want that… Also, using the octave clef does not do anything different in the cantai voice, stays om the same pitch.
2.2 still refuses to render any of my existing projects so is pretty useless. Only works with new ones (after a fashion). So I’m going to wait as well until there are clear signs of improvement.
I feel the same way. This was not ready for full release. While I applaud Richard for all his hard work, doing a full release and charging full price for it was premature. If I had to have something immediately, I’d go with something that’s much further along. I’m still on the fence, however, waiting to see. For me, I need a phoneme editor, particularly that supports IPA.
I must admit I feel a bit cheated. If it had shown signs of being a stable product I would have happily bought in early at half price and joined the beta. Now an unfinished product has been released I’ve lost that opportunity.
Perhaps all is not lost, as a very recent email from Richard confirms that the Sibelius beta will also include a full licence for Dorico:
“Sibelius beta opens mid-May. Launch is May 30. Preorder is $150 — lifetime licence, covering both Sibelius and Dorico.”
It seems odd not to have timed the Dorico and Sibelius VST releases to coincide, in order to avoid confusion over pricing. Why would anyone pay full price for the Dorico version when the Sibelius version - which also supports Dorico - is currently available at a 50% discount?
I own and enjoy using Synth V, but I’d happily purchase Cantai if I had more confidence in the product. Now that it has been released, it’s unfortunate that so many reports present a negative view. If it truly fulfils the promise of NotePerformer for the voice, I’ll happily dig out my credit card again.
It’s nice to work with Cantai. I feel this project is very promising.
What I especially need in Cantai:
-
At least pitch bend support for microtonal writing.
But ideally, it would be great to have proper native Dorico’s microtonal support, similar to how HALion handles it. -
Glissando support similar to NotePerformer.
And ideally, in a perfect world, glissandi without octave restrictions. -
Work with no vibrato mode.
Thanks, and all the best with your project!
well exactly. Those of us who use Synth V realise what the competition is – and Richard knows and has even said so himself on the Discord forum. A product which integrates seamlessly into Dorico would of course be a big advantage but we seem to be some way from that – even without taking into account the difference in quality and flexibility of most of the vocal models. But I will also happily dig out my credit card when Cantai starts to fulfil its promise (though I’d need higher standards than NotePerformer without NPPE)