Review: Japanese Hologram Pop Star Hatsune Miku Tours North America
arstechnica.com
arstechnica.com
That is really cool.
Anonymous-type groups posing in 3d avatars comes to mind.
I remembered that at the time,I found it implausible that it was possible to go online in Neuromancer without connecting to a fiber or something and that it was just handwaved away. He was right and I was wrong, it was not important.
"I mine for 100-year-old jeans" https://news.ycombinator.com/item?id=10284638
So today I will see John Kasich speak in San Jose and then I attend a J-Pop vocaloid concert in the evening. A most weird day even by SF Bay standards.
I attended Gaga's Washington DC show, had never heard of Hatsune Miku, and had no idea what was going on. Hatsune had live, human back-up dancers. By the end of the set I realized this could be the future of music - when I got home I devoured every piece of information on vocaloids I could find.
Lady Gaga talking about Hatsune: https://youtu.be/7QV63XHoenU
Reality truly is as strange as fiction!
From my perspective, it seems strange that anyone enjoyed the "real" thing in the first place.
For a more western example Toy Story 3 might have been a moving story, but it's no single persons vision given form. Instead it's the passion from many distilled into it's most marketable form.
I don't really get why you are comparing it to Toy Story.
Yes, it's voice sounds interesting, but it completely lacks emotion. So, the show ends up even more fake than Animation which still relies on traditional voice actors.
Most Vocaloid songs become popular in part because of videos that were animated by a single (usually different) person.
As for their voices lacking emotion, that's just, like, your opinion, man.
First of all, the songs are the main part. All of the rest is just secondary. You can not talk about anything related to Vocaloids without talking about the voice synthesis software, the voice banks and the songs using them.
Secondly, most animations that may or may not accompany songs are little more than slide shows with moving elements on the screen. Many songs don't even have that. Many only have a picture.
Thirdly, out of the songs that become very popular, some get fan made choreographies. In this process, motion capture using a Kinekt helps a lot, but you need to do fine tuning for a smooth result.
Each of these steps is within the realm of what a single very dedicated person can do.
Also, there is no such thing "the voice is synthesized from several people and effects" . Each voice bank has a single donor. And you can purchase the Vocaloid software and voice banks and use them without restrictions.
Sure, if you want to make a concert and charge for tickets, you might do the choreography part more professionally and use fancy projection technology, so people actually have a reason to pay for those tickets. Otherwise people can just compile a concert themselves and project it in their homes. I did just that for my family to explain the phenomenon to them.
As for your last paragraph, I don't think I can convince you something fictional is not "fake". You shouldn't expect animation to look real because that is not the point of animation. As for lacking emotion, I bet you have not heard songs that try to do just that instead of being gimmicky. Try this: https://www.youtube.com/watch?v=3BFvN-idN1s This was made by just two people. One did the music and lyrics, the other the illustration and video.
There is an intresting recent study that used really short clips of people laughing and asked: "Are they friends?" This is just laughter and it was played for people around the globe. People did better than random. Not 100% by any means but well past random change for a few seconds of laughter.
Now, extend this emotional void to a full song and it get's down right creepy.
As for the study, I believe you are referring to this: http://newsroom.ucla.edu/releases/friends-or-not-laughter-re... I have not read the article but I listened to the sound clips. They are definitely different. But I do not associate either with friends or strangers. If it were just these 2 clips I would probably identify them correctly because one is more involved and energetic and I know there is one of each kind. But I would probably find it more difficult to identify the type of each clip from a set without any extra knowledge.
That is a highly subjective judgement call, and i think i even know why it appears like that to you. Different languages express emotions differently, and particularly japanese appears "flat" to english people, to the point that english learners of japanese are taught to speak japanese like a robot to sound somewhat accurate.
Luckily there are some people making english songs with vocaloids as well. Have a listen to those and see if you get more out of them:
https://www.youtube.com/watch?v=cQKGUgOfD8U https://www.youtube.com/watch?v=7Skij_GE9n4 https://www.youtube.com/watch?v=XmifuQvFu8M https://www.youtube.com/watch?v=gfQShttQTd0 https://www.youtube.com/watch?v=D1xicOvYd3Q
The closet thing I could think of was how being readers can say words without understanding their meaning. The stresses follow the difficulty they are having not the content of a story. If you hear them speaking you can follow their emotional responses, or hear the words, but doing both set's up dissonance. Listen to 0:40 to 1:00 here and I think you can get what I mean. https://www.teachingchannel.org/videos/guided-reading-introd...
PS: While I don't think this is any more subjective than body language, the negative reaction might go away if I listened to a lot of this stuff. A speak and spell quickly goes from odd to just that's how it sounds.
Making an english song with english vocaloids is tricky to begin with due to the differences in language favoring japanese heavily for synthesis. Making an english song with a native japanese vocaloid is even harder, and the result is what you hear in "Lie". (You can also contrast this with "Last of Me", which was made with a recent upgrade on Luka, V4X, who can speak english and works out better; and of course the three Gumi songs, since Gumi is a natively english vocaloid who does quite well.)
In short: Yeah, you did accurately identify a foreigner speaking english slightly awkwardly.
As for the getting used to, yes, you get used to it, however that's not because you get used to vocaloids themselves, but because you get used to japanese songs, which feature vowel frequencies, vowel distribution patterns and rhythms in expression that are flat out alien to english speakers. This works out as "exotic and interesting" to some and "weird and off" to others. To give a more clear example, try and see if this song, which to to speakers of languages with similarity to latin vowels (germanic, slavic, chinese, japanese) sounds flat out beautiful, but is mostly "meh" to americans i found: https://www.youtube.com/watch?v=qCDoifoKkLA (Also, as for rhythm, it would probably sound even more weird for you if you knew the words, since many of her emphasises go on the japanese equivalents for "of", "too", "from", "?" and "!".
And, as long as people get value out of that, I'm okay.
You're forgetting one very real thing, music is more than just an aural and visual experience. There's a very good reason people shell out hundreds on tickets, drive 6 hours, wait in line for another 3, buy $8 beers, and fight a crowd of sweaty, amped up folks to get to the very front.
Nonetheless, the projection technology behind this is interesting.
I'm not sure pointing them out here would be useful.
For starters, cs702's comment here ( https://news.ycombinator.com/reply?id=11602173&goto=item%3Fi... ) is plain wrong.
It is wrong in the assumptions it makes. For your questions in particular.
- Yes, this was a concert (with paid tickets), in the US, attended by a large group of people.
- Yes and no, Miku is often called a Virtual pop star but that is a misnomer if by virtual pop star you mean anything else than just a fictional character who is a popular singer. It is just like saying Harry Potter is a wizard.
- It depends on your meaning of create. Miku was indeed created by Crypton. But what they created is nothing more than a voicebank for the vocal synthesis software Vocaloid, given it a name and commissioned someone to create the box art for the product. Everything else about Miku (her personality, likes, dislikes, hobbies, friendships, rivalries, clothes, the songs she sings, etc.) was made by fans. Miku is by no means controlled by anyone. Crypton, who have the original trademark for the character have given everyone the right to create what they want with the character. Through the fan art you make (songs, animations, fan fics, etc.) you have as much control on Miku as anyone else. There is no canon version of the Vocaloids.
Vocaloid is the vocal synthesis software created by Yamaha. The software uses voice banks usually provided by third parties. Each voice bank usually has a character associated. These characters are collectively called the Vocaloids. Hatsune Miku is one of the characters for one of the voice banks. It has nothing to do with all the visuals. Fans just created 3D animation software for crating animations with the characters. These animations were eventually used to create the concerts. Crypton, being the original trademark holder started providing these concerts to fans who requested them using fan made songs, fan made costumes and fan made choreographies.
- I leave the last question about the strangeness of reality to you.
Creating a new Vocaloid voice still requires a live singer, who must sing a long list of phrases. It's not a direct voice synthesis. Full singing synthesizers exist, but as yet are not as good.