If I'm working with a VB and am suddenly overwhelmed with OPINIONS, I shall post here! == Like I am now! So, now I'm gonna talk about my new sweetheart. [size=1.45em]DEMO[/size] ãUTAUé³æºé å¸ãã¿ã¤ã¬ã¼ã©ã³ãã¤ã¸ãç°æãã®ãACT3ã This is Nagi Haika (ç°æãã®). She's a japanese-language CV UTAU, whose production is based in Hong Kong. She belongs to a group known as "ChinaUTAU". I'll describe her voice as soft, but present. Her voice is mature, but not thick. She isn't HQ but I think she does many things very well, as I'll go on to explain. [size=1.45em]INCLUDES/CONTENT[/size] She has 2 pitches, with a prefix map. Her ground pitch is E4, with a pitch change up at A4. She has a full japanese-language syllable set, including diphthongs and consonants. She has romaji filenames, and kana aliases. She has 2 inhales, named like "br1" or "æ¯2". 2 is quick, while 1 is more casual. She has 6 ending exhales, named like "ab-" or "ãb". Her library is well-rounded and flexible. The VVs were a particular godsend, as they lessen the fiddling around time on a UST. The aliasing was through, including considerations to alternate romanizations and the prefix map. I'm not fond of prefix maps, but it felt very appropriate here. Her â pitch-change is softer than her ground pitch. It gives an effect like, get back Jojo. I never made much use of the ending breaths. I fiddled around with an "ob-" for a little while, before deciding that it wasn't worth it. I'm unsure whether the VVs and Cs were meant multilingually, the Cs in particular. Rather than being desonarants, we have an orphaned consonant. You can try using the phoneme [ s] for "desu", but it might be preferable to supress a [su] instead. I'll note that I remember running into one or two hiccups, like a recording that was never OTO'd, or a case of preutterance < overlap. I'd still describe her as very solid. Overall, she's very easy to work with. The aliasing and includes make her very flexible. She's very good for making a base UST, especially if you have trouble reading kana or working with UTAU in general. [size=1.45em]QUALITY[/size] Typical note length is ~half a second. Background ambiance is strong with this one. Pronunciation is not always on par. As I noted before, she is not an HQ bank. She's otherwise been recorded well, but the ambient noise can be intrusive if mishandled. Additionally, her pronunciation is consistent, but not always spot-on. "sh" is pronounced as "s", and the "f" and "h" are strong. I'll leave it at that, since that's as far as my linguistic experience permits. [size=1.45em]CHARACTER DESIGN[/size] Anime-style illustration, with suitable proportion/highlights/shading. Colours are primarily blue/red. It's an okay costume, but it doesn't really look like anything. I can't see any purpose in her accessories, or any relation between them -- it's just this cosplayer-fighter-student mishmash. The silhouette has some distinguishing contours, but the pieces aren't anything new. Other than that, I'd say that blue/red is a difficult colour scheme. It usually seems to work out well enough, but you have to be careful with accents if you want the illustration to look together. So, her design isn't quite garden variety, but it doesn't have anything really special or telling about it. It's fine enough in terms of recognizability, but I'd like to see something more together. [size=1.45em]RESAMPLERS/FLAGS[/size] resampler.exe: Quick strings can give a sound like chewing on plastic. However, background noise has the least influence here, this is Nagi at her strongest. fresamp.exe: REALLY brings out the background noise. At least it's natural background noise? haha. EFB-GT and tn_fnds: Seem to accept her. She becomes a little breathier. bkh01: Makes her even breathier! It's nice. TIPS: Balances the background and her voice evenly. I prefer the resampler.exe over the others, but I'd say that bkh01 and TIPS were the most promising alternatives. I never used any flags with her. When preparing your own cocktail, be mindful of the background noise. [size=1.45em]IN CONCLUSION[/size] We're all into UTAU for different reasons, and we all get something different out of it. So, if you're looking for a perfectly natural sound, then Nagi is just...alright. But if you like a robotic aesthetic, then I think you'll find Nagi a beautifully usable, unique addition to your collection. 'cause, being of the latter inclination, that's what happened to me. Ha. By the way, it looks like she's released a VCV since I first downloaded her. Considering all that stuff I just typed, I'm not sure she's suited to it...oh well. If you go to ChinaUTAU's youtube, you can see her "Leia" demo. Have a good day, everyone.