Cantai problems

Hi all,

Here’s what I have after an afternoon of futzing with Cantai. I put up the notation. Then the audio. This thing needs a lot of work. I’m not sure where to go from here.

And here’s the audio.

All I can say is YIKES!

more or less what I’d expect. If you’re not going to set text, you might as well use a choir that sings “Aah”'s properly – as I do on the rare occasions I write something like this.

I set some actual text in a different test as well. The word ‘business’ came out as bus (like the vehicle) nes.

I don’t want to be a negative whiner. I guess I’m just looking for some encouragement that this will get better as it goes.

YIKES! is right! Still, I will download Cantai today and have a go with it.

Joel

It’s such a good idea that I just want to help it become useable. The VST is glitchy right now, but seems to work, eventually. Tip: if the voices aren’t sounding, close it and reopen. That helps sometimes.

Does anyone else have any tips or thoughts?

My thoughts are that they have already got a lot of feedback on the beta. I would wait for the next beta release before diving in and hopefully many of those issues will be fixed.

Sounds like a plan. Give them time to sort it out.

in the newsletter, it says there will be at least one more beta before launch with as many bugs as possible quashed. We’ll see! Additionally they want to achieve “the pure-toned, no-vibrato choral sound so many of you have been asking for”. It’s a long way from that at the moment but I’d like to be optimistic. After all, Cantai has the huge advantage over SynthV or Ace Studio that it’s fully integrated into Dorico. Bugs, aside, the voice modelling is the main thing needing to be worked on

… which also (in?)advertently advertised Dorico’s next maintenance release.

indeed --if it actually is to be released on that day.

Chorus! That’s what I read too. And that’s exactly what I need, although I’m actually writing for. four-voice treble ensemble, not a full chorus, but I’ll take it, as long as we can get it in tune.

I’m not meaning to attack you, but I feel compelled to repeat what I sated on another thread to everyone freaking out in general: you are beta testing. If you don’t want beta results, don’t beta test.

to be fair, normally one would expect a beta released just two weeks before the release date to be close to the finished product – just with a few bugs to sort out. A lot of us don’t see Cantai as yet having reached that state.

Agreed. But a two-week beta period is kinda short. At this stage, it seems it needs a few betas to work out some things. I’d also hope that they would offer a free trial period after release, since the one language I need (Spanish) for the one work I would consider for Cantai won’t be there for at least a month or two. I think by pushing folks to get in on a 50% discount that expires in (as of now) 11 days, that changes expectations and it’s natural to assume this is a close to finished product. Clearly that is not the case, and it’s usually the process that one releases a public beta that has most of the functionality one needs with the caveat that some things (hopefully minor ones) may not work exactly as intended. Releasing a beta too early into the wild produces exactly what we’re currently seeing in this forum as well as on their Discord threads: people are finding tons of issues (which is good in some ways) but that will also discourage purchases since it doesn’t seem to be as feature-rich (only one language, no volume/balance, no way to tweak pronunciations, etc) or as reliable as one would have expected this close to an official release.

In terms of choral voices, I haven’t yet tried Cantai on a work (no lyrics) that I composed into Dorico last year. But the results with Noteperformer were pretty good IMHO, and so I made the album available on Bandcamp, Spotify etc. Here’s an excerpt:

I’ll be curious if similar results occur when I try it with the Cantai beta.

@donaldskoog1
I know you are frustrated but as soon as I listened to the audio, I burst out laughing uncontrollably! It sounded like each voice was in a different key.

I don’t really know a solution as I’m not beta testing but I did want to thank you for sharing your plight. I laugh with you and not at you!

I hope you find a solution soon.

Like I said, I’m not complaining. I’m just putting my problem out there. I’ve never beta tested before, so I don’t know what to expect. Some of the other responses seem a little more understanding than yours, so I’ll take yours in the spirit of tough love. This is giving me a better understanding of how things like this work.

That’s nice. I’d better look into Noteperformer as well.