Naysan♤

Eyyo this is MLo7, mostly post about vocal synth stuff and sometimes singing,,,, still trying tho

[Naysan♤ @NaysanMLo7]

@gnloop Gotta give them some credits 😭

[Naysan♤ @NaysanMLo7]

I think their engine works well with softer vocal like Baire here kinda eats a bit. But yea, most vocals suffers from having the generic feels to their voice. Idk, it might be an engine issue cus that specific in-the-nose feeling kinda noise is like very obvious in solid vocals

[ボーカロイド - VOCALOID7 @vocaloid_yamaha]

Omnivocal is now available!

"Omnivocal Basic" is a plugin included with #Cubase 16, powered by technology developed through VOCALOID.

For even more creative possibilities, "Omnivocal Pro" offers 10 voicebanks, Vocal Stack, and the full Omnivocal feature set.

Get Omnivocal Pro from the Steinberg Shop :
https://t.co/kJFlSNsXyN

[Naysan♤ @NaysanMLo7]

@wormwood_prod Hi bestieeee

[Naysan♤ @NaysanMLo7]

Aaaand became e2e model
Andromeda is doing so fineee
Training takes long as hell tho and the model is still severely undertrained 😭

Art by wormwood_prod

[Naysan♤ @NaysanMLo7]

Been brewing up a vocoder and architecture update. We are using auto-encoder and frequency domain neural vocoder instead.

Music: Mili
ustx: scoliosin

once again mixing sucks but well this is a quick test

[Naysan♤ @NaysanMLo7]

vocoder is somewhat undercookedso oop

[Naysan♤ @NaysanMLo7]

Been brewing up a vocoder and architecture update. We are using auto-encoder and frequency domain neural vocoder instead.

Music: Mili
ustx: scoliosin

once again mixing sucks but well this is a quick test

[Naysan♤ @NaysanMLo7]

Using nsf hifigan on a tiny update with crosslang TEST

Accent transferring implemented, but why would I want to hear Ritsu sing in English with strong Japanese accent lmao so im using English accent as a test

music & svp: FLAVOR FOLEY

mixing sucks but well this is a quick test

[852話(hakoniwa) @8co28]

「JIZURA　字面」
公開しました
歌詞から文字PV風動画を自動で組み立てるブラウザアプリです。

作例を添付します。こういうものが作れます
是非試してみてください。

https://t.co/9YTUeWg33k

[Naysan♤ @NaysanMLo7]

Only just started making instrumentsl wip and I already put cupcakke in it

[Naysan♤ @NaysanMLo7]

Like, if frequency domain discriminator generalizes into better scores than time domain, would that also means the same to the generator itself? Hmm. But the risk in phase incoherence is higher that's for sure

[Naysan♤ @NaysanMLo7]

Im amazed in how frequency domain neural vocoder solidifies the frequency details way faster than time domain, but both it still take as long to refine into a similar fidelity to one another. Though, I don't even know if different domain has different fidelity ceilings

[Naysan♤ @NaysanMLo7]

I wonder what could beat openau whisper features for ASR task to perform auto labeling... its encoder is really good

[Naysan♤ @NaysanMLo7]

Been taking a look closer at synthv2 outputs, and I am amazed about how their residuals are so naturally non-coordinated. comparing it to nsf hifigan, they seems to be similar but different. Almost as if sv2 excitations are nonlinearly generated while nsf hifigan is shaped...

[Siant(汗汁たくま) @siant36635755]

第5人格有識者の方教えてください！
これは次のシーズンに追加される新サバイバーですか？
このキャラでバッチ取りたいです！

[Naysan♤ @NaysanMLo7]

Even tho

[Naysan♤ @NaysanMLo7]

@VectorK91442 Yes actually lmao

[Naysan♤ @NaysanMLo7]

day and a half of training vocoder. I think it's doing quite alright so far?

[Naysan♤ @NaysanMLo7]

F0 shift (+12/-12) and taking away harmonic/noise

[Naysan♤ @NaysanMLo7]

recomputation is cool

SLabler 0.1.4

[Naysan♤ @NaysanMLo7]

When they say cipher machine prime but im still decoding

[Naysan♤ @NaysanMLo7]

Lowkey I think it has a bunch of useless features for a labeler lmao cus we have a pitch editor for funsies :sob:

[Naysan♤ @NaysanMLo7]

SLabeler has been slowly updated
Plan has changed, and I will only make one version instead of separate lite/full version.
I don't want to keep up with two of the same source code because one is already big enough.

https://t.co/1ctp004JeZ