To the people saying "Jerma needs to play Choicer Voicer":
NOOO!
Version 0.4 is crusty and rough. I dread this game blowing up before I can release an overhaul. That kind of publicity would just raise my blood pressure.
Vinny's stream was great, showed me some problems I hadn't even caught myself, but I guess that's how playtesting goes. That said, IF everything goes according to plan, I hope to release The Choicer Voicer within one (maybe two?) weeks.
The next update for v0.4 will finally give the studio packs their intended purpose: loading your custom glTF files as the studio.
I'm sure yours will look better than my beach here.
Full disclosure since I've been kinda quiet: I've been moving, and the whole process has slowed development substantially. As such, I think by now you all deserve a list of my ideas and hopes for version 0.5 and beyond. This is isn't everything, but it's a lot of it:
When debugging, I grab bits and pieces from peoples' packs and scramble stuff around for testing.
Then I occasionally gain lucidity, look at my screen, and wonder "what the hell am I doing with my life?"
I've made a video where I talk about my current progress with The Choicer Voicer, how some of the systems work, and what things I'm still figuring out:
I apologize that the game is still unreleased. For people who want to make and/or preview pack structure, I've uploaded the defaults here for reference.
I assumed users would've coalesced around a site for packs by now, but it hasn't happened. It's either Itch, the fan Discord, or google drive links.
Discord isn't a long-term solution because it's not readily available and Discord is AWFUL for archiving. Suggestions, anyone?
Here I thought my cumulative tweaks made my scoring algorithm better, but I'm watching this video of a streamer playing The Choicer Voicer and... why does it keep giving 4/5 EVERY time!?
Time to take a break from ver 0.5 to work on that headache again.
With the sheer amount of options I want to implement to customization for ver 0.5, it means I'm spending a lot of time just plotting it out.
It needs foundation repair, so with the base code, I have to do it all over again. And I want to get the job done right.
I can't believe it took me this long to think of doing this. Added voice packs as a tab in the customization menu, where you can toggle packs as visible or hidden. Uses the same goofy-yet-simple method of appending the reserved prefix "_ignore_" to hidden packs.
I feel like I'm in the final stretch. Menus look like crap, but right now I just care about them working right. Last big thing is getting my audio scoring algorithm just right... Other than that, it's a lot of little QoL stuff, mostly regarding customization.
Minor update, 0.4.13b:
* Added a search bar for the voice packs
* Trimmed some pauses during gameplay to quicken the pace
* Added options to make rounds go faster by skipping certain dialog
* Added a slider for microphone input volume
Me yesterday: "Why are people saying my game's slimegirl host looks like 'Bao?' Who is that?"
Me today: "Why have I been watching Bao videos for two straight hours?"
Someone made a category for The Choicer Voicer on Nexus Mods, but no one uses it (Unsure how easy or hard it is to upload there).
Someone suggested GameBanana at one point, but when I tried making a category, I couldn't. Got told it already exists even though it doesn't.
Small update for version 0.4.12, as "things it should've already had":
* Version is now displayed on upper right corner
* Including support to show "author" of packs. For now, via a TXT file called "_author"
* Contestant responses are now properly cut off for new rounds
Oh right, progress update.
Judge packs will be able to have more than 5 judges. Each voice session will randomly pick five judges from the pack. You will also be able to bind an audio file and success screen to a specific judge. Still ironing out the bugs, which there are many.
Small update, version 0.4.13 released. Of note is a change to scoring to make 5/5s easier to get. Also, an experimental change to the player modpacks: audio for them now partially supports the upcoming v0.5 format, with more dynamic file assignment.
Just posted my first bugfix with version 0.4.10:
* Fixes a bug where host dialog would return to the default
* Prevents a scenario where the game would crash if you made a pack with an illegal folder name
* Shows the current allowed length range when given the "too few" error
Call me a string theorist, cause I'm adjusting arbitrary variables until I get the outcomes I'm looking for.
(Coding this stupid exhaustive interface took me over three hours)
Quick pause in development to learn HTML. My right hand isn't a fan of the excessive use of "<" ">" and "/".
On a related note, I have a rudimentary neocities site for The Choicer Voicer. Here's the page on making voice packs for v0.4:
At least one person has informed me that their audio devices do not work with the current release. I have yet to diagnose the cause.
The customization menu free download includes the mic test, so I would suggest testing that before purchasing, to be safe.
The game will likely need some trial and error to properly use your custom studios. I am not well-versed in GLB/GLTF, but it seemed the best choice due to its versatility.
I've added a reference to give you a head start while I continue working on 0.4.14:
Continuing on data management, added the ability to hide clips. The reserved prefix "_ignore_" will prevent the game from loading it, and in the Hide Clips menu, you can easily hide any clips you've seen so far.
Also re-added a little tutorial from the previous iteration.
However, the next thing on my to-do list is rewrite the waveform box that records and displays your performance. Many people have reported issues with different mic delays, and in order to get the ability to account for variable delay, I need to completely remake that code.
My buttons. Simple innocuous little guys, right? WRONG. These spaghetti-coded monsters took me over five hours to untangle this weekend. On the bright side, you will now be able to set separate colors for button idle and button hover.
Status update, while I'm at it: The past week was mainly spent rewriting Crow-Sourcing, a microgame I made to test Twitch integration in Godot. Once I've confirmed the new version works, I can start adding Twitch voting into The Choicer Voicer.
While the previous version handled multiple voice packs just fine, I stupidly didn't do the same for the customizable settings. Any new packs for menus/judges/ect would overwrite each other.
Now, you can even swap between custom menus on the fly.
The game tracks your seen clips by their absolute file path, so if they're moved, it's not recognized. The "Rebind Seen Clips" option will find the changed paths and assign them, as long as all clips have unique file names.
I originally wanted to have local multiplayer working before releasing early access, but I'm realizing I didn't properly prepare for it with this iteration's structure.
I'm going to make sure the file system is future-proofed properly, and I'll see where it goes from there.
@rapscallionDlux
I needed more than just 3 or 4 people privately testing it. It was worth it because I've learned a lot. I knew going in that the first public release would be rough, but what I hadn't expected was this level of interest in it this quickly.
The author name only shows up on voice packs for now, but you should include it anyway in other packs you create for when the game loads those too. In version 0.5 it will likely merge into the configs, but I can make it backwards-compatible.
@ethan_josephson
Yes, each pack type is a user-customizable element. There is a local folder that you drop your pack folders into, and they all show up in the white box seen in the center screen.
@Omegpha
Thanks for the tip! 0.4.12 introduced an option that's supposed to make it generally easier, so that might've helped. But even then, it's just a stopgap I threw in; I need to try making something better from scratch
@Madclown55
Any image will be rescaled to 512x512 when it is put on the main screen, so anything larger than that is probably unnecessary. A lot of the images I put into packs are actually 256x256 in order to reduce file size.
@EmiruFin
...that's a good idea that I didn't even consider. Not sure how exactly I'll implement it just yet, but it should be doable. Thank you for the suggestion.
@NezomiFloof
There actually is no fullscreen option at the moment. It hasn't been worked on in a while, but when I was, it was incredibly glitchy for some reason. I'll have to look into it again.
@OfficialNoName9
I've got local multiplayer working, which I'm hoping will act as the base for online eventually. It's definitely the most daunting goal for me, but is important to have. I'm planning to try tackling it once I have a few other planned features implemented.
@brycenby
I think a big contributor is the buttons. I coded a custom shader in order to get that kind of look. Also the fonts; I use Waukegan LDO a lot. I would recommend collecting screenshots of UI elements that you want to emulate and finding the similarities between them.
@John_Kpoop
Ironic timing, considering I was just adding character limits in version 0.5 an hour ago. Spillover is funny, but I'll have to see whether I clamp the visuals for that or not, lol
@EmiruFin
Good question! That is definitely a limitation at the moment; I plan to enable JSON support for voice packs. In that way, you can have whatever number of images, and then assign images to voice clips via the JSON file.
@JayyyCoal
Since all versions use the same local directory to load packs from, every version will load the same settings and packs without any additional setup.
@lcfmetropolis
Online multiplayer will not be a feature in the early access launch; I have no experience with Parsec, so I wouldn't know if it would be a suitable substitute, but it would require transferring other players' audio to have VERY low latency in order to work.
@TechaTecha
I actually just posted an Itch announcement that talks about the mic issue, so if you've found a temporary workaround, that would be a good public spot to mention it for now:
@TorNis7
Currently, yes they are, haha. It's a temporary thing as I've been testing sound effects before having my musician create new ones. Before the ZE sfx, it also had sfx from the old e-Reader.
@EmiruFin
Not rude at all! Being able to assign images à la carte would be a nice QoL improvement, since yes, currently the only solution is to make a subfolder for voices that share a filler image. Although, if you have hundreds, assigning each image via JSON would be more time-consuming
@rapscallionDlux
It's a separate page on my Itch here:
I would just merge it with the main page, but then it'd be included as part of the purchasable game in order to access.