Search results

I’m a conlang producer, and the more pronounced accents sound like a terrible prospect to me.

According to the product page, it seems that SVS2 will drop Linux support. I’m extremely disappointed in that and am now considering SVS2 worse than useless to me.

Thanks to mahalisyarifuddin on Discord for sharing this.

Unfortunately doesn’t have a Linux version. I don’t think I’m likely to use it anyway, but this looks like an interesting program.

She sounds a bit breathy here? Never listened to the original, though. Hope more demos come out, maybe even one in a different language.

I don’t live in Japan, and I somehow wasn’t aware that there might be a delay between when I order through WorldShopping and when the order actually gets placed. Oops. :sad: Edit: got an e-mail telling me that the order’s been placed, so quite a relief.

What a madlad, using a then-unreleased voicebank in an MSTMS2 submission. Anyway, glad to see HARU coming out.

“Synthesizer U Tiny”?

I actually feel the opposite way about English phonetic input: I prefer X-SAMPA to Arpabet. (Totally has nothing to do with my other hobbies /s.)

AHS also put out a press release announcing that they’ll be involved in developing KAFU for SynthV and SEKAI for Voicepeak.

Simple: Tuning isn’t the hardest step of writing a song; the lyrics are. :P

WHAT. Why bring CeVIO to Linux when you can bring KAFU to SynthV? (I’m not saying this in a bad way.)

There are 2 characters: (by the file names in the repo) Laychie and Layney.

I’m still gonna say Herman Miller.

I predict that Animen will mess things up badly enough for Dreamtonics to intervene.

Just got back into this after a long hiatus.

Given the spacing of the letters in the character image, I think her name is more likely to be Keino, not Kyono.

I wonder if the multilingual support is going to allow mixing languages at the phoneme level. If so, then this is big for conlangers, or at least those who use Windows or macOS.

I’m not going to buy Yuma just yet, but he’s on my next-to-buy list. I want to get another AI voicebank to test pronunciation techniques on. In particular, I’m curious how he sounds when singing in English.

Some time ago, I discovered that lowering the Strength parameter to 20% on a stop phoneme with Eleanor Forte AI would make it sound like a fricative. This could be used to get an [x] when singing in English, or [θ] and [ð] in Japanese. Unfortunately, version 105 no longer exhibits this behavior...