tl;dr I'm doing reclist bullshit again. Namely, I'm taking that thing I've been working on for the past 5 years, the VCV English, and (thinking about) turning it into that thing I've been working on for the past 3 months, CV but for people who hate themselves like me. I'm trying to find a way to make it bearable to record and oto and simultaneously sound not terrible but it is incredibly difficult.
You know how I have this motherfucker:
This is just a semi-vcv thing (I'm officially calling it TCV-VC / Transition-Consonant-Vowel Vowel-Consonant)
And I have this other motherfucker:
This is the same concept but with cv + semi-vc instead of vcv (I'm officially calling this one CV-TC / Consonant-Vowel Transition-Consonant, I've done this w/ primarily Japanese + done some very minor testing in English).
And my thought process was "dang why don't I just like, take the second one's concept and, line-by-line, convert the first one to that?"
The plain and simple answer consists of two parts:
Consonant clusters in TCVVC aren't their own independent entity unless they're being used as endings, so it sounds weird and awkward to have to force it. There are several missing consonant clusters and no real consistent way to make each one sound good (nor is there good syntax for this imho). Adding the optional converters would make this, like, a non-issue, which is still a bad thing because it kinda breaks how this method is supposed to be used, but having options is nice.
Lastly, the reason TCVVC is so small (for what it is) is because of the diphthong converters. Rather than removing consonant combinations, I removed vowel combinations.
So the thing is, I really can't go and base another reclist off of this without adding what's missing. I can't go in line-by-line and recreate something that isn't there. I know that it's conceptually sound and all the work lies in duplication.
You know how I have this motherfucker:
This is just a semi-vcv thing (I'm officially calling it TCV-VC / Transition-Consonant-Vowel Vowel-Consonant)
And I have this other motherfucker:
This is the same concept but with cv + semi-vc instead of vcv (I'm officially calling this one CV-TC / Consonant-Vowel Transition-Consonant, I've done this w/ primarily Japanese + done some very minor testing in English).
And my thought process was "dang why don't I just like, take the second one's concept and, line-by-line, convert the first one to that?"
The plain and simple answer consists of two parts:
- The first thing isn't nearly complete. It's got some weird stuff missing, and a lot I wanna add (it's also a lot larger than I would like it to be but that issue can easily be solved by this thing I'm doing). I'm obviously not going to attempt to do any of this before finishing up the original method. That includes finishing up the Japanese version (which isn't huge so it's not a big deal).
- It's genuinely a lot of work, despite how easy I've made it sound, and I can't test how viable it is without making a whole voicebank, tbh (this is partially untrue, because technically the first method, outside of mid-only and actual VCs, is near-identical).
Completion
The current set of... stuff... missing from TCVVC is genuinely not a lot but it's enough to drive me nuts rn- Single-consonant transition-consonant endings (I have to make some hard choices with this one)
- VV / diphthong / liquid consonant converters (removing normal VC would make these vital, but they're already pretty important as it stands for increased realism)
- Missing consonant clusters
- Optional consonant cluster converters (for if I ever decided to give you the option of making your bank even tinier. It's already pretty small but that's a valid choice imho)
Consonant clusters in TCVVC aren't their own independent entity unless they're being used as endings, so it sounds weird and awkward to have to force it. There are several missing consonant clusters and no real consistent way to make each one sound good (nor is there good syntax for this imho). Adding the optional converters would make this, like, a non-issue, which is still a bad thing because it kinda breaks how this method is supposed to be used, but having options is nice.
Lastly, the reason TCVVC is so small (for what it is) is because of the diphthong converters. Rather than removing consonant combinations, I removed vowel combinations.
So the thing is, I really can't go and base another reclist off of this without adding what's missing. I can't go in line-by-line and recreate something that isn't there. I know that it's conceptually sound and all the work lies in duplication.
Comments
Post a Comment