Musings and opinions on the various Vocaloid & Vocaloid2 voicebanks
In no particular order,
Gakupo: I love him. Hated Gackt, though. I think I started out with the wrong songs for introducing myself to his music, so I'm not going to just write off his stuff, but all the same, I don't think I can stomach listening to his cover of the fantastic Paranoid Doll again. I ought to try investigating the effects of various settings on Gakupo's voicebank, though, since he mostly sounds the same everywhere. Did also find one song on NND where he'd been ported to Vocaloid3 -- I'm not sure I like what it does to his voice there. Trying to reserve judgement on Gakupo in V3 for later. Overall, I love the timbre and when the user has his voice break at certain notes and everything. He's quite probably one of the most easily realistic voicebanks out there, imho -- perfectly imperfect is totally the way to go. ♥ Plus, having him sing in english, when done right, is just awesome; his accent is really nice to listen to. I hope they do up some expansion packs for him -- giving him a greater range of styles would be nice; apparently his bank is sampled from Gackt singing at a more mellow/soft tone than not. His voice can be a lot more powerful than the voicebank seems to be able to pull off, so it would be interesting to see what people come up with, if they ever do manage to sample the gamut for him. Even if they don't, I'll happily keep Breathe on my iPod playlist forever. (it's on YouTube somewhere too, if you want to see the PV with english subs.)
Luka: Her voice itself is fine, but her consonants are so soft it's often hard to listen to her alone, especially when she sings in english, and still suspend disbelief that it's a person and not a program. If she did better at pronouncing those, I think I'd like her more, but as she is, she sounds good in duets with Gakupo (like Tsugai Kogarashi) and in other songs where she's mostly singing in Japanese (like Stardust Utopia, Blackjack, etc)
Piko: I really didn't care for him at first. I really didn't. Another high-pitched voice, like some ten-year-old boy. Not that that's bad, but I'm not fond of the high, clear voices (which is one reason I love Gakupo so much, I think -- his is neither!) so I didn't bother looking into his stuff much. And then I was browsing on the VocaloidOtaku forums for something, and came across someone's post where they were trying to get his voice to go lower... apparently adjusting the Gender Factor (basic pitch formant control setting) by about 25 does the trick, and makes him sound... like an adult, or at least an older teenager. And oh wow, does he sound fantastic like that! I fell for him just from the one song, Chaining Intention, they posted as an example! But I can't find others where he's been pitch-adjusted like that. It seems most users like using him as a kid. Sigh.
Len: Much the same as with Piko, overall -- didn't care for him, and then I heard a particular song, called Transmit, where they have him singing with a pitch-change of... what was it, 35? and he sounded more like a teenager. They had his original voice singing as a backup for him in the background, and it's probably my favorite song using him. (Butterfly On The Right Shoulder (Migikata no Chou) is so totally better when sung by Gakupo. *nod*) Overall, I'd consider listening to his stuff when he's been pitched down, but otherwise, I'm not really much of a fan.
Rin: Don't get me started. Another high-pitched voice, bleh. I'm sure she's great for those that like that, but it's really not my cup of tea, you know? Like... at all. Pass.
Miku: Same as with Rin, except she's freakishly popular and I can't seem to get away from her. Someone please shut her up! If I wanted to hear yet another high-pitched, clear female singer, I wouldn't be ignoring the radio stations. I can acknowledge that Miku's bank is very high-quality, yes. Her phonemes blend well, she can be made to sound very natural, etc -- but I just don't like her type of voice.
Gumi: Her voice, on the other hand...! ♥ It's not so crystal-clear, it's humanishly imperfect again, a little touch breathy/hoarse. And I love that quality in it. I'm finding more songs on NND using Gumi's Whisper and Sweet expansions that I'm falling so in love with. Sayonara Midnight has to be one of my favorite Gumi songs ever. You wouldn't be able to tell it wasn't a human singing, if you didn't already know, I'm sure... my sister didn't catch on when I played it for her! Haha.
Kaito: I actually hated him at first -- the phonemes were chopped up too obviously, and he sounded like he swallowed a synthesizer, or was just someone playing with a Vocoder and an audio editor; something about his sustained notes just screamed "fake!" to me. I'd avoided his songs just as much as I avoided the UTAU banks' songs. But I've since found a few songs where he's been used right, and even more of his bank in V3, and one where they used VocaListener on him, and... I am in love! He sounds fantastic. ♥ There are a plethora of different takes on his Thousand-year Solo Song (Sennen no Dokusou Ka/Uta) and I've yet to find one I didn't like. The SPTM mix of Pane Dhiria is fantastic as well. Plus, there's one artist who did some amazingly odd and sometimes wonderful things with his voice, like the cool and a touch creepy Scissors, and the astoundingly catchy dance/techno song, Infinite Loop.
Sonika: Currently, I don't like her. I don't hate her, but I don't like her. She sounds like a higher-pitched Luka -- really understated consonants, an odd timbre to her voice that makes it sound muffled, somehow, and to top it all off, her vowels all sound quite open to me (Japanese-style pronunciation.) I'm tempted to give her a try, to see if I can do something with her that'll redeem her to my ears, like lowering her pitch (maybe she'll sound more realistic then? Adult lady, rather than being so... cartoony?) and maybe even having her sing something such that she sounds like she has a Japanese accent, which seems to be able to excuse quite a bit of mispronunciation.
Prima: If she's being used correctly, I love her. The problem is, I rarely hear her being used correctly. Everyone wants to use her for pop songs, it seems like. I would absolutely love to hear her do a cover of the Lucia di Lammermoor, but until then, singing a West Side Story song isn't bad! And Who didn't expect Ave Maria?
Tonio: Same as with Prima; when he's used well, he's fantastic. I've found him used to sing Yesterday and Billy Joel's Just The Way You Are, and in duets with Al as backup, to sing Scarborough Fair and with Sonika for a really good rendition of Thriller. Unfortunately, I also keep finding him used to cover Lady Gaga songs, of all things. That shit needs to stop. Like, immediately. Having Gakupo cover Brittney Spears is hilarious, to me. Tonio doing Bad Romance? Shudder! KILL IT WITH FIRE.
Big Al: Mixed bag, here. On one hand, I love his voice, and I've heard him used really well many times! Like for Yesterday, Golden Slumber, Bruno Mars's Just The Way You Are, Carrying You, and Salamander to name just a few. On the other hand, listening to more songs with him has helped to make me more aware of English pronunciation, and how the sounds actually used differ from the letters of the words. Like all the times when the user has Al end a word with the "t" sound ("what" in the middle of a sentence, or "that") when Al should've used the "d" sound, for example. I can't seem to un-hear all those little flaws, now. Augh! Long story short, I love Al, but I really wish people who use him would pay more attention to the reality of english in practice, rather than on paper, since Al's pronunciation is so clear that those little slip-ups are really very noticeable. :( And hearing him sing in Japanese is just... LOL. His accent is horrible. ♥ (It doesn't sound so bad sometimes, though! like here where he does a duet with Leon.)
Sweet Ann: It's high time I got around to hunting down more stuff using her, I think. I've heard a couple good ones, like Night Illusion and Green Day's Boulevard of Broken Dreams but I haven't made up my mind on her yet entirely. I do like that she sounds like a mature lady rather than yet another teenager, though, so that's already points in her favor in my book. (Biased? Me? Naaaaaaaaaw. ....okay, maybe a bit.) So I guess I'm leaning toward liking her, but I'd like to hear more stuff done with her in general, and at her natural range, not up so high as in some songs I won't even point to.
Leon: He sounds about as synthetic as one might expect one of the two very first Vocaloid banks to sound, but I have to say I do like him. I've heard a couple songs already where his fake-ness was used to great effect, either hidden or emphasized by the music -- the techno, house, dance, electronica, etc sorts of genres work really well for him, in this regard. Try a Daft Punk cover, Digital Love, or even better, I Still Want U. I definitely need to go hunting around for more of his songs done like that.
Lola: The other very first Vocaloid bank out there, released at the same time as Leon was, by the same company and everything. She's a grown lady with a deeper voice -- and I like that. So far, someone's had her cover Cher's Believe quite nicely, and I like her cover of Owl City's Fireflies better than Owl City's version. (I think I've heard the original too many times.)
Miriam: I haven't really heard enough with her to be able to tell what I think of her yet. Her voice is sort of on that line between "kinda high" and "just right" the way I've heard people using her, as well as that odd line between "realistic enough" and "too synthetic" -- but I suspect that like with Leon, it'll really depend on how she's used. I'd like to hear her doing some techno/dance stuff, I think. Until then, the best song I've found with her is her doing Donna Burke's Wondering. While it's pretty well done, I think it's still a little high-pitched for my tastes, and I can tell she's a program. Hmm. Edit: Wow. Just found one I love! Sunset.
Meiko: I love her. Another lady with a lower voice -- released before Kaito was, actually -- but sadly not half as popular as the younger-sounding Crypton voicebanks. Miku needs to step back a bit for Meiko's Kowase Kowase, Forgetting Suicide, Human Genome, Chiisanasekai and definitely for her awesome duet with Kaito, Ban Kogarashi. Especially for being a first-gen Vocaloid, she sounds quite nice to me. ♥
Kaai Yuki: I love this kid. Disco Chocolat is just adorable, and her rendition of Castle On A Cloud completely blew me away. She needs to see more use. She really, really does! ♥ (And I don't mean with more Lady Gaga covers. Yes, I'm looking at you, people on YouTube. Quit that already! Not all of us want to hear endless Vocaloid covers of those songs! :P)
Kiyoteru: A rare one, to be sure, and he does sound a little bit like Kaito -- he covers the same range, and his timbre isn't as wildly different as, say, Gakupo's or Tonio's or Al's -- but he's still quite distinct to my ears. Guilty Verse has quickly become a fave of mine. I wish I could find more good stuff using him solo....
Gachapoid: Heh... not really to my tastes so far, but I bet someone could make him sound like Donald Duck and do some awesome stuff that way!
Lily: Not really to my tastes either. Her voice is nice enough, but too strong for me. I like Lily Lily Burning Night okay, I guess, but I'm not always in the mood for it.
VY1, VY2: Can't say much about them; I haven't heard enough yet.
Iroha: I kinda lol, since I'd heard the name and went looking for a song to see what this one sounded like. I found a nice version of Wishing On The Same Star that has Iroha as the lead singer, and I actually thought it was a male voice at first. Yeah, I laugh now, but at the same time, I really like that sort of androgynous quality to her voice.
SF-A2 Miki: Satellite was in one of the albums I picked up. I rather like it, but at the same time, she sounds enough like Miku that I'm not in a hurry to go hunt down more of her songs yet. A more listenable version of her, given, but still similar enough that I'd rather hunt down more of Leon's techno instead.
UTAU banks: No. Just... no. There's one of Lucifel from El Shaddai, and that one's good for a laugh, but UTAU is just so much lower quality than Vocaloid, that it's generally extraordinarily grating to me. To each their own; I can't stand Teto/Ted no matter how many people think they're so wonderful.
...
...*head-scratch*
I know I'm leaving out some banks, but I'm pretty sure they're Vocaloid3 ones, and haven't been out long enough for me to catch much of them. Ah well. Later, perhaps!
This entry was originally posted at http://ladyasul.dreamwidth.org/20892.html. You can comment there using OpenID.
Gakupo: I love him. Hated Gackt, though. I think I started out with the wrong songs for introducing myself to his music, so I'm not going to just write off his stuff, but all the same, I don't think I can stomach listening to his cover of the fantastic Paranoid Doll again. I ought to try investigating the effects of various settings on Gakupo's voicebank, though, since he mostly sounds the same everywhere. Did also find one song on NND where he'd been ported to Vocaloid3 -- I'm not sure I like what it does to his voice there. Trying to reserve judgement on Gakupo in V3 for later. Overall, I love the timbre and when the user has his voice break at certain notes and everything. He's quite probably one of the most easily realistic voicebanks out there, imho -- perfectly imperfect is totally the way to go. ♥ Plus, having him sing in english, when done right, is just awesome; his accent is really nice to listen to. I hope they do up some expansion packs for him -- giving him a greater range of styles would be nice; apparently his bank is sampled from Gackt singing at a more mellow/soft tone than not. His voice can be a lot more powerful than the voicebank seems to be able to pull off, so it would be interesting to see what people come up with, if they ever do manage to sample the gamut for him. Even if they don't, I'll happily keep Breathe on my iPod playlist forever. (it's on YouTube somewhere too, if you want to see the PV with english subs.)
Luka: Her voice itself is fine, but her consonants are so soft it's often hard to listen to her alone, especially when she sings in english, and still suspend disbelief that it's a person and not a program. If she did better at pronouncing those, I think I'd like her more, but as she is, she sounds good in duets with Gakupo (like Tsugai Kogarashi) and in other songs where she's mostly singing in Japanese (like Stardust Utopia, Blackjack, etc)
Piko: I really didn't care for him at first. I really didn't. Another high-pitched voice, like some ten-year-old boy. Not that that's bad, but I'm not fond of the high, clear voices (which is one reason I love Gakupo so much, I think -- his is neither!) so I didn't bother looking into his stuff much. And then I was browsing on the VocaloidOtaku forums for something, and came across someone's post where they were trying to get his voice to go lower... apparently adjusting the Gender Factor (basic pitch formant control setting) by about 25 does the trick, and makes him sound... like an adult, or at least an older teenager. And oh wow, does he sound fantastic like that! I fell for him just from the one song, Chaining Intention, they posted as an example! But I can't find others where he's been pitch-adjusted like that. It seems most users like using him as a kid. Sigh.
Len: Much the same as with Piko, overall -- didn't care for him, and then I heard a particular song, called Transmit, where they have him singing with a pitch-change of... what was it, 35? and he sounded more like a teenager. They had his original voice singing as a backup for him in the background, and it's probably my favorite song using him. (Butterfly On The Right Shoulder (Migikata no Chou) is so totally better when sung by Gakupo. *nod*) Overall, I'd consider listening to his stuff when he's been pitched down, but otherwise, I'm not really much of a fan.
Rin: Don't get me started. Another high-pitched voice, bleh. I'm sure she's great for those that like that, but it's really not my cup of tea, you know? Like... at all. Pass.
Miku: Same as with Rin, except she's freakishly popular and I can't seem to get away from her. Someone please shut her up! If I wanted to hear yet another high-pitched, clear female singer, I wouldn't be ignoring the radio stations. I can acknowledge that Miku's bank is very high-quality, yes. Her phonemes blend well, she can be made to sound very natural, etc -- but I just don't like her type of voice.
Gumi: Her voice, on the other hand...! ♥ It's not so crystal-clear, it's humanishly imperfect again, a little touch breathy/hoarse. And I love that quality in it. I'm finding more songs on NND using Gumi's Whisper and Sweet expansions that I'm falling so in love with. Sayonara Midnight has to be one of my favorite Gumi songs ever. You wouldn't be able to tell it wasn't a human singing, if you didn't already know, I'm sure... my sister didn't catch on when I played it for her! Haha.
Kaito: I actually hated him at first -- the phonemes were chopped up too obviously, and he sounded like he swallowed a synthesizer, or was just someone playing with a Vocoder and an audio editor; something about his sustained notes just screamed "fake!" to me. I'd avoided his songs just as much as I avoided the UTAU banks' songs. But I've since found a few songs where he's been used right, and even more of his bank in V3, and one where they used VocaListener on him, and... I am in love! He sounds fantastic. ♥ There are a plethora of different takes on his Thousand-year Solo Song (Sennen no Dokusou Ka/Uta) and I've yet to find one I didn't like. The SPTM mix of Pane Dhiria is fantastic as well. Plus, there's one artist who did some amazingly odd and sometimes wonderful things with his voice, like the cool and a touch creepy Scissors, and the astoundingly catchy dance/techno song, Infinite Loop.
Sonika: Currently, I don't like her. I don't hate her, but I don't like her. She sounds like a higher-pitched Luka -- really understated consonants, an odd timbre to her voice that makes it sound muffled, somehow, and to top it all off, her vowels all sound quite open to me (Japanese-style pronunciation.) I'm tempted to give her a try, to see if I can do something with her that'll redeem her to my ears, like lowering her pitch (maybe she'll sound more realistic then? Adult lady, rather than being so... cartoony?) and maybe even having her sing something such that she sounds like she has a Japanese accent, which seems to be able to excuse quite a bit of mispronunciation.
Prima: If she's being used correctly, I love her. The problem is, I rarely hear her being used correctly. Everyone wants to use her for pop songs, it seems like. I would absolutely love to hear her do a cover of the Lucia di Lammermoor, but until then, singing a West Side Story song isn't bad! And Who didn't expect Ave Maria?
Tonio: Same as with Prima; when he's used well, he's fantastic. I've found him used to sing Yesterday and Billy Joel's Just The Way You Are, and in duets with Al as backup, to sing Scarborough Fair and with Sonika for a really good rendition of Thriller. Unfortunately, I also keep finding him used to cover Lady Gaga songs, of all things. That shit needs to stop. Like, immediately. Having Gakupo cover Brittney Spears is hilarious, to me. Tonio doing Bad Romance? Shudder! KILL IT WITH FIRE.
Big Al: Mixed bag, here. On one hand, I love his voice, and I've heard him used really well many times! Like for Yesterday, Golden Slumber, Bruno Mars's Just The Way You Are, Carrying You, and Salamander to name just a few. On the other hand, listening to more songs with him has helped to make me more aware of English pronunciation, and how the sounds actually used differ from the letters of the words. Like all the times when the user has Al end a word with the "t" sound ("what" in the middle of a sentence, or "that") when Al should've used the "d" sound, for example. I can't seem to un-hear all those little flaws, now. Augh! Long story short, I love Al, but I really wish people who use him would pay more attention to the reality of english in practice, rather than on paper, since Al's pronunciation is so clear that those little slip-ups are really very noticeable. :( And hearing him sing in Japanese is just... LOL. His accent is horrible. ♥ (It doesn't sound so bad sometimes, though! like here where he does a duet with Leon.)
Sweet Ann: It's high time I got around to hunting down more stuff using her, I think. I've heard a couple good ones, like Night Illusion and Green Day's Boulevard of Broken Dreams but I haven't made up my mind on her yet entirely. I do like that she sounds like a mature lady rather than yet another teenager, though, so that's already points in her favor in my book. (Biased? Me? Naaaaaaaaaw. ....okay, maybe a bit.) So I guess I'm leaning toward liking her, but I'd like to hear more stuff done with her in general, and at her natural range, not up so high as in some songs I won't even point to.
Leon: He sounds about as synthetic as one might expect one of the two very first Vocaloid banks to sound, but I have to say I do like him. I've heard a couple songs already where his fake-ness was used to great effect, either hidden or emphasized by the music -- the techno, house, dance, electronica, etc sorts of genres work really well for him, in this regard. Try a Daft Punk cover, Digital Love, or even better, I Still Want U. I definitely need to go hunting around for more of his songs done like that.
Lola: The other very first Vocaloid bank out there, released at the same time as Leon was, by the same company and everything. She's a grown lady with a deeper voice -- and I like that. So far, someone's had her cover Cher's Believe quite nicely, and I like her cover of Owl City's Fireflies better than Owl City's version. (I think I've heard the original too many times.)
Miriam: I haven't really heard enough with her to be able to tell what I think of her yet. Her voice is sort of on that line between "kinda high" and "just right" the way I've heard people using her, as well as that odd line between "realistic enough" and "too synthetic" -- but I suspect that like with Leon, it'll really depend on how she's used. I'd like to hear her doing some techno/dance stuff, I think. Until then, the best song I've found with her is her doing Donna Burke's Wondering. While it's pretty well done, I think it's still a little high-pitched for my tastes, and I can tell she's a program. Hmm. Edit: Wow. Just found one I love! Sunset.
Meiko: I love her. Another lady with a lower voice -- released before Kaito was, actually -- but sadly not half as popular as the younger-sounding Crypton voicebanks. Miku needs to step back a bit for Meiko's Kowase Kowase, Forgetting Suicide, Human Genome, Chiisanasekai and definitely for her awesome duet with Kaito, Ban Kogarashi. Especially for being a first-gen Vocaloid, she sounds quite nice to me. ♥
Kaai Yuki: I love this kid. Disco Chocolat is just adorable, and her rendition of Castle On A Cloud completely blew me away. She needs to see more use. She really, really does! ♥ (And I don't mean with more Lady Gaga covers. Yes, I'm looking at you, people on YouTube. Quit that already! Not all of us want to hear endless Vocaloid covers of those songs! :P)
Kiyoteru: A rare one, to be sure, and he does sound a little bit like Kaito -- he covers the same range, and his timbre isn't as wildly different as, say, Gakupo's or Tonio's or Al's -- but he's still quite distinct to my ears. Guilty Verse has quickly become a fave of mine. I wish I could find more good stuff using him solo....
Gachapoid: Heh... not really to my tastes so far, but I bet someone could make him sound like Donald Duck and do some awesome stuff that way!
Lily: Not really to my tastes either. Her voice is nice enough, but too strong for me. I like Lily Lily Burning Night okay, I guess, but I'm not always in the mood for it.
VY1, VY2: Can't say much about them; I haven't heard enough yet.
Iroha: I kinda lol, since I'd heard the name and went looking for a song to see what this one sounded like. I found a nice version of Wishing On The Same Star that has Iroha as the lead singer, and I actually thought it was a male voice at first. Yeah, I laugh now, but at the same time, I really like that sort of androgynous quality to her voice.
SF-A2 Miki: Satellite was in one of the albums I picked up. I rather like it, but at the same time, she sounds enough like Miku that I'm not in a hurry to go hunt down more of her songs yet. A more listenable version of her, given, but still similar enough that I'd rather hunt down more of Leon's techno instead.
UTAU banks: No. Just... no. There's one of Lucifel from El Shaddai, and that one's good for a laugh, but UTAU is just so much lower quality than Vocaloid, that it's generally extraordinarily grating to me. To each their own; I can't stand Teto/Ted no matter how many people think they're so wonderful.
...
...*head-scratch*
I know I'm leaving out some banks, but I'm pretty sure they're Vocaloid3 ones, and haven't been out long enough for me to catch much of them. Ah well. Later, perhaps!
This entry was originally posted at http://ladyasul.dreamwidth.org/20892.html. You can comment there using OpenID.