Cantai Reviews
Cantai is a great feature that greatly enhances the music notation experience in MuseScore. However, I hope the 8 GB RAM requirement can be adjusted. My laptop has 8 GB physically installed, but Windows reports only 7.7 GB usable, causing Cantai to reject local rendering. I also hope the 180-minute weekly Online Processing limit can be increased or removed, or made more accessible to everyone. These limitations prevent users from fully enjoying this otherwise excellent feature. I hope the developers consider improving these requirements.
Of course, there are still issues, and there are more languages than English or Latin. But the difference with only oe’s, uu’s or aa’s is big.
it works very well with latin and english. but pourly with italian, where all the “e” sounds like a “i” . it is really disappointed in face of the huge song in italian. The future release does not mentionned italian. Help me !
Genuinely a very good product, happy to support! I would love to see the ability to solo out voices in the choir sounds using staff texts like "solo" and "tutti." Other than that small suggestion, this is great!!!
Dear developpers, what you have acheived is phenomenal, thank you! There are however some bugs, mistakes wich bother very much when you make a big score with vocalists. I’ve just made a 1st scene of an oratorio in latin for double choir, I’m sure that if I send it to you it would be very helpful to understand what things need to be resolved in first place (because there a real mistakes when you have 10 staves for choir, the sound desapears, some parts suddenly begin to sing in english instead of latin. dinamics don’t work ect ect) . Thank you very much again! It’s obviously the only library that can be used here in Musescore for now.
I tried this with MuseSounds Pro to test if I like it, and it's actually not that bad. That said, there's still some glaring issues. The first is the dynamics. The voices overpower EVERYTHING. Even when I had a full orchestra playing at fff, the vocals here would over power them. In some cases, the orchestra would overpower the voices. But still. Also, the voices themselves have wonky dynamics. Next, is the vocals themselves. Sometimes, they would incorrectly pronounce (or sing in this case) lyrics. I'm sure they're built for English and Latin lyrics because I tried them with a song I composed that uses Latin lyrics, and they turned out fine. But I'm sure it's for these lyrics only b/c I tried it with Indonesian lyrics and it didn't work.
Sounds are improving however render times have increased which is very annoying to deal with. Also, there are still problems with vocals like Karolyn missing notes or going sharp. Also the voices will randomly go loud at certain times which is also pretty annoying. Please fix these issues! Other than these major issues it is quite solid.
This is just amazing, how Good voices are, but sometimes, it does not detecte the Latin and it pronounces it as English. It needs to be able to read perfectly every single language it says it knows, because is so weird when my characters are speacking Latin they feel british and is so akward and weird.
I tried an earlier version of Cantai for MuseScore and found it a great help in making choir practice tracks. The updated voices in 3.0.1 sound even more lifelike, and I was excited to try them out. For some reason, the 5-10 minute rendering time per song from before has now jumped to several hours per voice part. After trying several suggested fixes, it's still crazy slow. An app with great potential, but not worth the stress and frustration it's causing.
This is a gamechanger. I wish I could get instrumental stuff sounding as good as my vocals now.
I started to write this review while on my 15th minute of waiting for a 3-bar passage to render with four voices, again. This has been consistently happening since 3.0.1, despite numerous reinstallations. Previously this was a great tool for hearing ideas sung, and render times were on the order of seconds, even on my M2 Mac. Now, even on my primary Windows computer with great specs, I have to wait multiple minutes to hear even short passages. A GPU acceleration patch for S/D has been released that ostensibly improves speeds by 2.5x. Okay, but Ottetto voices were released months ago on S/D and they are just now coming to MS. Am I going to have to wait months to get a patch that makes the application I pay for actually usable?
The sound can be excellent. Not perfect, not complete, but really remarkable. Love it when it works. Too bad that I do not have local generation with my brand-new AMD PC and have to rely on often slow-ish, sometimes non-functional online generation, with a quota hanging over my head like a sword of Damocles. Generally speaking, the MuseHub/MuseScore combination is not very stable when it comes to MuseSounds, I "lose" them temporarly pretty often, installing them was a matter of persistance.
Love this software. As a church composer for my Samoan and English Methodist Parish, this software is helping me hear my hymns in English before I teach it. However, when it comes to the Samoan language, there is none. Is the Samoan language coming soon? Will really appreciate it if the Samoan language comes sooner.
About a couple of years ago, this program started being developed. It kept announcing that it was a program for opera singing. This year, when I finally decided to try it, it was still far from operatic in any way. I hated it, because even the choir voices were juvenile and amateurish and do portamenti. The pronunciation of Latin, which is evidently done by singers from the USA, is horrible. This is a new program. It's trying to sing correctly, in different languages, including all dynamics, articulations, etc. That is a massive achievement. They keep on working with it. Above all, they deserve to be given a chance, because the developers are listening to the community and constantly trying to improve.
I would give this program currently 2.5 stars. I am a choir director working with a small church choir. This sound allows me to make practice tracks for my choir which is a huge plus. However - sustained notes are consistantly off pitch. And often times the voices add random ordaments in the playback. I have the tenor line decentding chromatically (Think barbershop style). It is supposed to be from C-B-Bb, and it goes C-B-(C-B)-Bb, think the parenthisees as two grace notes. I was also kind of disappointed that the choir sounds are very ‘Pop’ instead of classical. Don’t get me wrong - I could use the pop sound for certain songs, but in most of my compositions I would prefer to have a more classical choir.
You cannot open the plugin to change language settings in Musescore, so everything is pronounced in English. I contacted technical support and received an automated response; no human interaction was possible.
Developer Response: Hi there, please see here for details on how to customize the language settings: https://handbook.musescore.org/customization/language You can also find more details on how to use Cantai within MuseScore Studio here: https://cantai.app/faq If you continue to have issues, please feel free to reach back out to customer support and we'll do our best to assist. Thanks!
theres no bass, and the song im using it for has a bass solo and the low notes just sound really bad with the baritone voice, i would love this if you guys would add a bass voice!
1. it doesn’t work with wi-fi turned off, which to me means it doesn’t really work offline. 2. it is still on v2.0.1 (updated 2 months ago), despite dorico and sibelius having already received the 3.0 update (july 1rst). I see no point in using a plugin that isn’t updated regularly - especially when the current version has such huge issues with vibrato and no SATB choir. i will buy a cantai license at full price once those 2 issues have been fixed for Musescore.
Great work. Sometimes it shows that it's work in progress, but nevertheless it will prove useful.
I bought this plug-in becuase in Dorico and sibelius versions Spanish is already supported and I thought that the MuseSounds version will also be updated in order to support Spanish. Also I need to customize jow many voices I want my choir to be made of and not getting a number of voices determined randomly by the plug-in. Is this going to be possible sometime in the future?
Developer Response: Thanks for your feedback! We’re actively working on improvements, and you can expect updates soon.
It's a good Idea, but, we should be able to set parameters like sound attack, vibrato, glissando...
If you think about using this for anything other than contemporary and modern choir arrangements, don't buy it. It doesn't sync with other instruments very well. It works for my purpose (contemporary a cappella) but can sound kind of weird, especially with higher and lower notes for each vocal type. It's certainly a cool concept that has a lot of potential. It will be much, much better in a few years or so I imagine, but this version is still cool to mess around with and somewhat useful.
I am mainly an orchestral composer, but I have written a lot of choral work as well. This fits a much more contemporary style need. This works in pop, jazz, modern-style application. It does not work so much as a traditional large choral CHOIR library. The diction is modern, microphone right at the voice, lots of “scooping” up to initial pitches. That all being said, the fact that it reads normally written lyrics and gets the pronunciation correct is amazing. The voices blend well. I use this currently to layer in the mix on top of using EastWest’s Hollywood Choirs, so the diction is better.. I am fluent in Votox, but it’s not perfect. This smoothes out the roughness. But it has to be deep in the mix otherwise it stands out. Looking forward to a proper larger choir version of this….
Ok so, this will be a bit long. So, i'm a composer. I write mainly orchestral music, and liturgical Catholic music, mostly in latin and spanish. (I live in México tho.) I have acces to Cantai because i'm paying the Muse Sounds Pro subscription. This context is pretty much important because, as a professional musician this product is COMPLETELY UNUSABLE. If you're not a prof, you can end here: Try it. It sounds decent, it Works mostly well in english, the voices are ok. It might help your choir to know what to sing. But if you're a professional, keep reading. 1. The "voices" are MEEH. Like, meh. There is one soprano that it's barely usable, but it tends to sing like "pop". This definetly does not work if you're doing classical, but for "pop-ish" sounds, it might work. The devs should put a lot more attention into how the sound actually is, as "singing - playing styles" it's one of the most important things we as composers have in account. Specially, because more than one "voice" adds grace notes that are not written anywhere in the score. Specially the men, when going from one note to another. The "baritones" are completely unusable, don't even try. The tenors, there is one that Works well as pop music, but all the other ones just don't work for classical. One even sounds like a woman voice tuned down. 2. In musescore, the processing SUCKS. That's the only way i can describe it. And it's not my computer, i have a pretty beast of a machine as desktop (i7, 32gb ram, 12 gb video dedicated, 2tb ssd etc.), and a MacBook Pro m3 24gb ram, and a pretty much stable 500mbps symmetrical internet connection and IT TAKES FOREVER TO PROCESS, and suddenly it stops and it says "Online sounds are not ready", and you can't do anything with it. Like, anything. Just totally unusable. And real deal here it's that IT WON'T TELL YOU WHY. It just stops, and Good luck finding why it won't process. There is not an error log, or "click here to retry", or "you should change this syllabe, or anything. NOTHING. And there's no offline processing, so either you open the file again or try to move something in the text, but you won't ever know why it won't render. This should be easy to fix: a freaking toggle that lets you "Process offline" or "Process online", because i understand that this processing takes quite a lot of resources, but my brother, i have them, let me use it. 3. The "Choir" mode it's actually 3 or 4 voices of the same "soprano, alto, tenor or baritone" (there is no actual bass.), and it sucks. It sounds pretty awful. 4. The dynamics are PAINFUL. They don't even try. That's the only thing i have to say. 5. About the languages, well, it says clearly that they only support english. It DOES NOT SUPPORT LATIN. Latin does not work as expected. You should have a language option, that you select when selecting the voice in the mixer. So, conclussion: If you're not a PRO, you should try this. Pretty nice tech, it "Works" within musescore. If you're a pro, i STRONGLY SUGGEST DO NOT PAY FOR THIS STANDALONE. It's not even close to be worth it the full price. It might be in the future, but not now. Suggestion to the devs: Check out "cantamus.app". They've been making this kind of synthesis since 2022, and it's pretty stable and predictable. Thank you, and good luck.
Developer Response: Hi there, thanks for getting in touch with us. Could you please share more details by submitting a support ticket here? Our team will be happy to help: https://support.musehub.com/hc/en-gb/requests/new