-
If you do not recieve your confirmation email within a few hours, please email haloutau@gmail.com with your username for manual validation. Your account should be activated within 24 hours.
You may also reach out via any other listed contact on Admin Halo's about page: https://utaforum.net/members/halo.194/#about
This is interesting! It sounds great and the transitions seem smoother than normal ARPAsing- is this a modified reclist?
I could see this being an entry-level voicebank for professional music producers/an alternative to the Dreamtonics mascotless VBs. From the pro musicians I’ve spoken to, they all seem a bit turned off by the mascots.
I could see this being an entry-level voicebank for professional music producers/an alternative to the Dreamtonics mascotless VBs. From the pro musicians I’ve spoken to, they all seem a bit turned off by the mascots.
It uses an original ARPAsing list called “LSE” (Lexical Simple English) that I plan on releasing alongside DB-SVS. As its name implies, it’s a word-based list that utilizes only single-syllable words from Ogden’s Simple English dictionary, and it’s made to be as easy to record as possible for a word-based list. That said, on a technical level, it’s a pretty ordinary ARPAsing reclist, and only uses standard ARPAsing phonemes, with additional add-on lists for [zh], [q], and [dx] as well.
The smooth transitions are mostly the result of the specific style of configuration, which I intend on making a guide for, and providing a toolkit with some executable Python scripts to make it easier. Part of it may also be the particular recording style used for DB-SVS too, since it was meant to be very consistent, and lacking a lot of the accent-specific quirks of typical American English that may often make synthesizing English more difficult.
Last edited: