can someone give me feedback?

[ammamanagucci]

sooo it's been a hot minute since i've used UTAU, all i do is like, open up USTs, work on them for an hour or so then give up and move on. but i've been really dedicated to covering don't stop me now by queen using teto english (vsqx is messy but i'll make this damn thing work) and i was wondering if anyone can tell me how it sounds so far?? here's a preview. it's not tuned, it just has basic portemento and vibrato and that's about it, and i didn't mix any of her voicebanks together, it's just her english. thank you!! (harmonies are gonna be tei, momo, ritsu, and teto)

 

[LemonSquareLyricSquares]

Hello! Before I get started on saying my thoughts, I just want to say that I checked out the preview on Soundcloud and I'm so sorry to hear somebody gave a negative, unhelpful review. I personally didn't find the cover "trashy" at all, and I'm very glad that you decided to seek help for making it better!

I also wanna preface this by saying I personally don't find myself to be too good with UTAU myself yet despite being in the community for a couple of years, but some people say I'm really good with English UTAUs for some reason?? So I thought, "eyyy I can share my knowledge about this," lol.


That being said!! Here's my feedback! (Sorry in advance for the length in the long version; it's nothing too bad, I'm just going in depth, lol!)

Spoiler: The Long Feedback

Lyric Choice
So, to start with the positives first, the lyrics you translated from the VSQX to something Teto can read sounds pretty good! I know that's kind of an odd thing to compliment, but trust me, when I was starting out using Teto even with her TTENglishHelper plugin, the @ sounds always made her sound weird and I never thought about changing them because that's just what the plugin suggested, lmao. While you're doing pretty good with it now, just keep in mind certain sounds like @ can be replaced with sounds like A, O or even a (although [a] is less likely to have ending sounds, like I don't think for example [a t] exists-) to better fit the lyric! This can also apply to i sounds by replacing them with E.

Secondly, I won't be commenting on tuning because you said only basic portamento and vibrato exists for the moment, and that's a-okay! So, what I want to give out advice on next is how the voice flows. While I can still understand Teto okay at some points, her voice sounds rather choppy and hard to understand at others. Other times, her voice sounds a bit stretched, like in the "m" in "there's no stopping me". There are some solutions to this, however, which I'll get into now.

Help/Consonant Velocity (beta) + STP

Firstly, select CTRL+A and go to Note Properties. Then, next to Preutterance and Overlap, select Clear if the boxes are greyed in. If they're not, then feel free to ignore this. Then go down to Consonant Velocity (beta). Consonant Velocity is basically just the speed of the consonant, for example, the "k" in "ka". I looked up the BPM for Queen's song and it's BPM is 156, which is pretty fast in my opinion. "Normal Speed" for consonant velocity is considered 100, but feel free to experiment and find out something that works! Just, unless you're doing a super fast song or something, I heard it's best to not go past 200, lol.

Next, same on the Note Properties page is STP! STP will basically subtract milliseconds from the offset, or start of the oto, hence the name, "starting point". STP, for me, goes hand in hand with Consonant Velocity because while Consonant Velocity speeds up the consonant speed, STP allows for more room to be present in the offset, for which some room may be needed in the cases of weird OTOs or just weird USTs/VSQXs in general that already have some defaults set by the creator, you know. Feel free to tell me if this doesn't make any sense. >-<

Again though, feel free to experiment with this value as much as you want. It's really all about experimenting. Lol.

Crossfading

After you've gotten all that done, click OK and go to the top right corner, where your crossfade options are. Now, P2P3 and P1P4 are the actual crossfading methods, while ACPT just manages to figure out what kinds of overlapping needs to be done. When it comes to my UTAU covers in English, so far, I've been doing ACPT > P2P3> ACPT but this could be a really outdated method or only used for Japanese vocals, who knows. Also Optimize, the fourth option, is supposed to fix Envelope errors, although I'm not quite sure myself exactly what it means by that, for example, if it means the yellow exclamation mark over the note whenever there's an error, then that can easily be fixed within Envelopes themselves, so. >->

Modulation

Finally, maybe erase Modulation or at least turn it down. I'm not sure what it's set at but Teto sounds a bit weird in some places and that could also be a factor.

Sorry for this huuuge wall of text explaining what I need to explain. If it sounds complicated, then feel free to ask me questions and everything, while I'm not an UTAU Guru or whatever (heck no lmaO) I'll be glad to try and help in whichever way I can!

Spoiler: TL;DR Version because I got carried away lol

1. The notes you chose that Teto's English could read was great! For somebody who hasn't used UTAU in a hot minute, you're good with knowing how to work with inputting her English lyrics, lol.
2. Maybe try to work on lyric flow, as in, mess around with Consonant Velocity (which is Consonant speed, basically) and STP (starting point).
3. Also try to play around with crossfades. P1P4 and P2P3 are the only crossfade options, whilst ACPT (I don't know what this means, lol), OPT (optimize) are more for before or after crossfading.
4. Since Teto sounds strange in some places, if Modulation in Note Properties is set or greyed out, then maybe erase it or enter in something lower if you know what it is, as too high of a number causes the UTAU to sound odd.
5. Experiment! Have fun!

Spoiler: One more thing about Teto to make this smaller

Also, you said at the end you didn't mix her voicebanks. And that's fine! There's nothing wrong with that at all. But just so you know, there's a way to combine all of Teto's voicebanks into a mega voicebank in which you can use both Japanese and English sounds together, thus making it easier for her to show emotion without dreadfully going back and forth and rendering different vbs. You can do this by simply extracting her unzipped vbs into UTAU and clicking "Yes" to the pop-up windows. If you already knew this, then that's great! I never knew it until very recently though, lol, so I just thought maybe you'd like to know as well.

Again, so sorry for the wall! There are probably some tutorials that explain this all way better than I can, but I didn't want to be one of those people who're like "Oh, this needs work, here's a tutorial". Like, that's not a bad thing to do, really, but I dunno, I feel doing this is more fun and helpful.

Just so you know, although this was a huuuge post, I don't think your cover was bad at all. Like for one: I get carried away, lol. And also, UTAU English can be hard, lmaO. Even I can't grasp all of it yet, but that doesn't stop me from wanting to help in any way I can, lol.

You have a really nice cover with some great potential in your hands right now, and I'd love to see where it goes in the future! I honestly would love to see it once more of its once it's done, as using different vbs for Harmonies like you said sounds really intriguing! I'm happy to hear you're dedicated to covering this! That being said, I hope this helped!!

 

[ammamanagucci]

gah!! thank you so much for this, i never, ever mess with STP or consonant velocity in my covers bc i never understood them?? but yes! i will go mess around with those and stuff. i'm not sure how to go about tuning this honestly, i think less is more, right? i rarely ever tune anything (which is kinda bad, i've been doing UTAU for 4 years lmaooo).

but yeah, teto's english is really easy tbh?? and i referenced how freddie pronounced things in the original song because most people just use the plug-in and call it good. tbh i can just play some music and quickly convert a vsqx into a UST for teto

as for crossfading, i just do P2P3 > P1P4 > ACPT (maybe it's short for accept? tbh i just followed an old UTAU tutorial from back in the day and always stuck with it, i just never knew what it was for.)

yeah, i have all of teto's voicebanks combined but tbh,,, this is what happened last time i tried mixing her voicebanks together and i'm not the best at tuning either lmao.

i really do appreciate the feedback, this is what i really needed before continuing this cover i'll definitely post updates and ask for opinions once it's fixed w/ tuning if you wanna follow the thread or whatever. we could also swap discords if that's easier.

 

[LemonSquareLyricSquares]

gah!! thank you so much for this, i never, ever mess with STP or consonant velocity in my covers bc i never understood them?? but yes! i will go mess around with those and stuff. i'm not sure how to go about tuning this honestly, i think less is more, right? i rarely ever tune anything (which is kinda bad, i've been doing UTAU for 4 years lmaooo).

I mean! There's nothing wrong with simple, less = more tuning if done right, you know? :0c If you don't know how exactly to go about tuning, maybe try looking up simple songs you like and listening to how the singer's voice fluctuates, then try replicating that in UTAU? Again, there's nothing wrong with simple tuning, but it takes some extra work to make sure it still sounds expressive and not flat, you know?

(Also, congrats on doing UTAU for four years! )

but yeah, teto's english is really easy tbh?? and i referenced how freddie pronounced things in the original song because most people just use the plug-in and call it good. tbh i can just play some music and quickly convert a vsqx into a UST for teto

Teto's English is pretty easy once you get the hang of it, yeah! And I'm glad that you think you're pretty good at converting VSQXs for her!

as for crossfading, i just do P2P3 > P1P4 > ACPT (maybe it's short for accept? tbh i just followed an old UTAU tutorial from back in the day and always stuck with it, i just never knew what it was for.)

Lmao, same with method? I've been using ACPT>P2P3>ACPT for years but I also never knew what it meant for a long time, lol. But yeah! You can always just play around with the options, and if you don't like it, reset both with the button on the far right and in the Note Properties with Preutterance and Overlap. ;v;

yeah, i have all of teto's voicebanks combined but tbh,,, this is what happened last time i tried mixing her voicebanks together and i'm not the best at tuning either lmao.

If you don't mind me saying, it sounds like the Modulation was up, there? I could be wrong, but that's what it sounds like. To be honest, I kinda like Teto's tone there, it's pretty interesting. And no worries, it's okay! Tuning's pretty tricky sometimes, so it's completely fine if you don't know how to tune things at times! I honestly get the same thing all the time, haha. :">

i really do appreciate the feedback, this is what i really needed before continuing this cover i'll definitely post updates and ask for opinions once it's fixed w/ tuning if you wanna follow the thread or whatever. we could also swap discords if that's easier.

Of course! It's no problem at all! I can totally continue following the thread until you're finished with it; whatever you want to do is okay with me!

 

[ammamanagucci]

If you don't know how exactly to go about tuning, maybe try looking up simple songs you like and listening to how the singer's voice fluctuates, then try replicating that in UTAU?

that's what i'm gonna try to do, but it's kinda hard to do that bc i'm dumb and i may be tone deaf?? idk, maybe i can see if there's just his vocals and try to replicate that. also! i discovered that canon made an autotuner, so i'm gonna mess around with that and see what i like?? i may opt for basic tuning, it's better than no tuning at all. i'm sure mixing can help, right?

If you don't mind me saying, it sounds like the Modulation was up, there? I could be wrong, but that's what it sounds like.

yeah!! that's what another utau-er said, but i think it's because i mixed sakibi and her english, and it's a pain in the ass because her english is more mature bc she voiced it recently, whereas her japanese banks are from when she was about 15, which causes a huge difference in her voicebanks, and that's why i'm not gonna be mixing her banks with this cover. i may revisit dancing queen (god i love that song) and fix the modulation and whatnot, especially the tuning, but for now i wanna focus on this one and see what i can do. but yeah, tuning is hard and i don't wanna ask someone else to tune it :<

but i really do appreciate your feedback! i'm playing around with the auto tuner and i kinda like it?? we'll see about it, though. it's messing with the consonant velocity and the STP, though, so i'm just resetting those to default. thank you, again!