Korelis Labs

Building foundational software for autonomous security & quantum reliability. Creators of RegenX, QSHL, RegenX//OS & DeadKey.

[Korelis Labs @KorelisLabs]

Working on Callio's Vocal Lab. Users will be able to understand their voice better, built a model of their voice and license it out if they choose!

[Korelis Labs @KorelisLabs]

Callio now understands vocal range, this weekend we are Hopefully going to get real human vocals mixed and mastered....🙏🤞

[Korelis Labs @KorelisLabs]

@Latitude10Tech we are trying to take it to new level...we shall see lol

[Korelis Labs @KorelisLabs]

Here's some Kawaii Metal made by Callio. How do we rank compared to suno?

[Korelis Labs @KorelisLabs]

@TinMadDog26 @araseb_ It's not public yet... you can see more about it abs sign up for the beta on our website. We are hoping to have the beta ready by December....Hopefully

[Korelis Labs @KorelisLabs]

@lightsilver323 Callio AI Music Platform - You actually control the composition, stems and individual instruments, keep what you like and change what you don't, and even use your own vocals.

[Korelis Labs @KorelisLabs]

@blackjaguar011 @araseb_ if you got any questions or suggestions, let us know.

[Korelis Labs @KorelisLabs]

More control. Suno is great at generating a finished song from a prompt. Callio lets you actually work on the song, so you control the composition, individual instruments/stems, keep what you like and change what you don't, and use your own vocals. It's closer to AI assisted music production than a generate button.

[Korelis Labs @KorelisLabs]

if it was just another prompt to AI song generator, I'd agree lol

Callio is different. You actually control the composition, stems and individual instruments, keep what you like and change what you don't, and even use your own vocals.

Suno generates a song for you. Callio is being built to help you make the song.

[Korelis Labs @KorelisLabs]

Callio beta sign ups are live on our website! Please note: we are still months away from a good beta... but progress is progressing 🤣

[Korelis Labs @KorelisLabs]

We're building an AI music platform that doesn't treat a finished song like one giant black box.
Callio composes the music, exposes the actual stems, and lets you regenerate individual instruments or sections without throwing the whole song away.
Lock the drums.
Rewrite the bass.
Change the guitar.
Replace the vocal.
Keep everything else.
We're getting very close to putting this in people's hands.
Here's the current desktop build 👇

[Korelis Labs @KorelisLabs]

Here's a look at Callio! Our AI Music Platform, still in development but we are really close to user testing!

[Korelis Labs @KorelisLabs]

Callio's backend is now working with minimax. We can generate very high quality sounds significantly faster than before. SoulX is the next goal! #Callio #AIMusic #Suno

[Korelis Labs @KorelisLabs]

Here's a demo of Callio's audio quality. we don't have a time line of when Callio will be ready for first beta test, but if follows us, we will keep you posted so you can be one the first to try it out!

[Korelis Labs @KorelisLabs]

We have been working very hard at Korelis Labs over this last year, and with the addition of our new co-founder and lead engineer, we’ve also been making a lot of progress on our AI music program, Callio. Unlike @suno, our engine is trained on music theory and composition, not stolen works from artists. Demos coming soon.
Would you actually use an AI music tool that wasn’t trained on existing songs? Or is that a non-issue for you? #AI #AIMusic #music

[Korelis Labs @KorelisLabs]

Callio, our music theory and composition engine took major strides this last week in development. We are fine tuning sub-genres and vocals. You'll be able to eventually liscense your own voice! Stay tuned #ai #aimusic #suno #audio #callio

[Korelis Labs @KorelisLabs]

We've been pretty quiet about this project, but it's finally at the point where I want to start showing what we're building.

This is Callio - an AI music creation platform we're building at Korelis Labs.

The idea behind Callio is different from the typical "type a prompt ... get a finished song" approach.

We're building it around an actual music production workflow.

Callio generates the underlying composition and musical structure, then gives that structure to separate rendering systems for instruments and vocals. The goal is to keep songs editable and give creators control over the music instead of hiding everything behind a single generation button.

We've now built the core Studio workflow, integrated multiple generation/rendering backends, added AI vocals, and we're working toward cloud vocal rendering so using Callio won't require owning a powerful GPU.

There's also a much bigger idea behind the vocal system: voice identity should belong to the person who owns it. We're designing Callio so vocal models can eventually be created, controlled and licensed by their owners rather than treating someone's voice as just another model baked into the platform.

We're not ready for public testing quite yet. The remaining work is mostly about connecting everything into a reliable end-to-end experience, packaging it, improving project persistence and making sure someone can actually install Callio and create music without knowing anything about the infrastructure underneath it.

But it's getting close.

This is a demo from the current build. We are only 2 people, please be patient with us 🙏

More Callio updates coming as we work toward the first tester release. #AI #AiMusic #MusicSoftware

[Korelis Labs @KorelisLabs]

We’ve been quietly bringing a Nokia Lumia 1520 back to life at @KorelisLabs.

Mainline Linux is running, hardware enablement has come a long way, and right now we’re deep into bringing up audio.

13 years old. EOL. Still not dead.

Hey @Nokia… want to see a Lumia resurrected? 👀📱🐧

[Korelis Labs @KorelisLabs]

Since my last Lumia 1520 update, this has gone far beyond simply booting Linux.

Mainline Linux now cold boots reliably with display/backlight control, USB networking, SSH, battery reporting, suspend, RTC wake and physically verified light/proximity sensors. The frozen baseline passes 17/17 regression tests from a cold boot.

Last night brought the next major breakthrough: the Lumia’s WCD9320 audio codec was physically enumerated over its ADSP-controlled SLIMbus, and its hardware identity was successfully read as 0x0102/0x0001.

Both fresh initialization and adoption of an already running codec passed cleanly, no bus errors, timeouts, duplicate devices or fallback recovery.

Audio output isn’t working yet, but the codec, transport and real hardware communication path are now proven.

Old Windows Phone hardware still has life left in it.
https://t.co/dDdEwOklZV⁠ #windowsphone #Linux #postmarketOS