Forum - View topic
Answerman - Why Are Anime Voices Recorded After Animation Is Done?


Goto page Previous  1, 2, 3  Next

Note: this is the discussion thread for this article

Anime News Network Forum Index -> Site-related -> Talkback
View previous topic :: View next topic  
Author Message
KutovoiAnton



Joined: 03 Mar 2013
Posts: 991
Location: Vladimir, Russia
PostPosted: Mon Dec 11, 2017 4:10 pm Reply with quote
HolsteinCrab wrote:

I'm also willing to bet it's the dubbing after animation process that explains why ad-libbing is such a cardinal sin in Japanese voice acting. The Mawaru Penguindrum Japanese commentary had a few anecdotes of seiyuu either accidentally slipping ad-libs into scenes (and apologizing afterwards) or thinking they managed to sneak one in and have it accepted only to be told after the take it had to be redone. Technically I'd say anime is visually a very forgiving style for adlibs because the lip movements are deliberately kept vague but a sudden ad-lib can still muck up the tempo of the scene or just not fit so I can see why some directors are very strict about that. Seiyuu in more comedic roles seem to be allowed more freedom in that regard though as Tetsuya Nomura commented how he let Shigeru Chiba throw in ad-libs as Kefka for Dissidia and Kenichi Suzumura for Demyx in Kingdom Hearts II - however those were games so different deal Laughing


Comedy series allow their voice actors to ad-lib quite a lot. Good half of Kazuma's lines in KonoSuba are ad-libbed. Osomatsu-san got a plenty of those as well. I imagine Gintama got a lot of it as well.


As for the voices being recorded before animation - it doesn't always mean a high budget. Sometimes it simply means that animation isn't ready yet, like in Dragon Ball Super Laughing
Back to top
View user's profile Send private message
Lord Geo



Joined: 18 Sep 2005
Posts: 3016
Location: North Brunswick, New Jersey
PostPosted: Mon Dec 11, 2017 4:30 pm Reply with quote
KutovoiAnton wrote:
As for the voices being recorded before animation - it doesn't always mean a high budget. Sometimes it simply means that animation isn't ready yet, like in Dragon Ball Super Laughing


Nice try for a silly joke, but Justin's answer effectively indicates that voice work is never done to finished animation. It's usually just basic pencil animation that hasn't been finished up yet, though some of it (usually small parts) is in time for recording.
Back to top
View user's profile Send private message Visit poster's website
Shay Guy



Joined: 03 Jul 2009
Posts: 2657
PostPosted: Mon Dec 11, 2017 5:06 pm Reply with quote
HolsteinCrab wrote:
Ufotable is an exception thanks to the insane budgets they manage to boast


Pretty sure that's a myth -- from what I've read, their deal is that they've got all their staffers under one roof like KyoAni, instead of the constellation of sub-subcontracting that defines most anime.

Lord Geo wrote:
Nice try for a silly joke, but Justin's answer effectively indicates that voice work is never done to finished animation. It's usually just basic pencil animation that hasn't been finished up yet, though some of it (usually small parts) is in time for recording.


See, for example, voice-acting scenes in Shirobako and especially Girlish Number.
Back to top
View user's profile Send private message
EricJ2



Joined: 01 Feb 2014
Posts: 4016
PostPosted: Mon Dec 11, 2017 6:14 pm Reply with quote
MarshalBanana wrote:
Akira and Redline had pre-recorded voices, however they had complex facial movements, so I assume there was noway around it..


Even in early 30's US cartoons, the Fleischer studios worked quickly believing that dialogue was handled the same way as Disney's Steamboat Willie made the transition from silent to sound cartoons: Animate the key-dialogue mouth movements, and let the actors improvise sounds and reactions to boost up the rest of the former silent-cartoon "pantomime" business.
That was the appeal of Betty Boop, and, more famously, Popeye, when Jack Mercer started muttering crazy asides when the character's mouth wasn't moving or his back was turned, and created "the first ad-libbing cartoon character" in the recording booth. (Watch a vintage B/W Fleischer Popeye cartoon with Mercer on the voice, and you'll know why they cast Robin Williams in the '80 movie. Laughing )

It was only after Disney, MGM, and the theatrical Warner cartoons had to keep up the "quality" image that they started taking the time to animate the dialogue for dialogue-friendly sound cartoons, and the rest of the NA industry followed suit.
Japan, OTOH, had to make TV anime from scratch, and never got out of the primitive early days, just refined it.
Back to top
View user's profile Send private message Send e-mail
mewpudding101
Industry Insider


Joined: 07 Apr 2009
Posts: 2212
Location: Tokyo, Japan
PostPosted: Mon Dec 11, 2017 7:46 pm Reply with quote
This confuses me. Because all the stories and experiences I know are different.

Usually, when recording in Japan for Japanese animation that has not yet premiered, the animation is NOT usually done. You’ll have a set of key frames for the actors to put their performances to, but the animation itself is not done yet.

I hate to refer to my own article, buuuut...

http://www.anime-now.com/entry/2017/02/02/000001

Quote the voice of DBZ’s Goku:

“…I apologize to anyone from Toei in the room, but the footage we’re given to go off of are lines. The pictures aren’t completed,” she said, “We all put our voices to the lines.”

While the voice acting I’ve done has had completely done animation, that’s only because it’s dubbing over the Japanese dialogue with an English script.
Back to top
View user's profile Send private message
relyat08



Joined: 20 Mar 2013
Posts: 4125
Location: Northern Virginia
PostPosted: Mon Dec 11, 2017 8:14 pm Reply with quote
mewpudding101 wrote:
This confuses me. Because all the stories and experiences I know are different.

Usually, when recording in Japan for Japanese animation that has not yet premiered, the animation is NOT usually done. You’ll have a set of key frames for the actors to put their performances to, but the animation itself is not done yet.


It's the order of the process, not the completeness of the animation, that is important here. In the West, you record lines and then you begin animating. In Japan, you storyboard, begin animating, and then record lines.
You're right, in many cases, probably even the majority these days, the animation is NOT finished. It is, however, already begun. That's the distinction.
Back to top
View user's profile Send private message My Anime My Manga
TheAncientOne



Joined: 06 Oct 2010
Posts: 1949
Location: USA (mid-south)
PostPosted: Mon Dec 11, 2017 10:33 pm Reply with quote
FilthyCasual wrote:
Soulwarfare wrote:
I got a question, why do some anime studios like to ruin their openings by adding some unnecessary sound effects?

Because it's the finale and finales are special.

Not always. I seem to recall Knight's & Magic doing that for several episodes. I also believe it hasn't been the only anime to spread the sound-effects in the OP over more than just the finale.

It is a bit of a strange one, however. When I went back to confirm my recollection, I initially thought the sound effects started with the OP for episode 11 (but they start only after the brief scenes from the previous episode, which was another new addition), but then I noticed episode 9 & 10's OP had sound effects, but only near the very end.
Back to top
View user's profile Send private message
leafy sea dragon



Joined: 27 Oct 2009
Posts: 7163
Location: Another Kingdom
PostPosted: Tue Dec 12, 2017 12:30 am Reply with quote
A corollary question:

Since animation before acting is the standard in Japan, are there many Japanese voice actors who are comfortable with performing before the animation? I'll bet they're pretty different experiences that require different kinds of practice and readiness, so I wonder if the majority of Japanese voice actors become accustomed to performing while watching the animatics (or storyboards or however far along they are) and actually get uncomfortable with having to perform without seeing anything.

Niello wrote:
His made me appreciate the voice acting in Land of the Lustrous even more. It is really noticeable how much better and more natural many VA sound when they don't have to put their effort into matching the lip movement.


The more I hear about that show, the more I want to see it (legitimately, before anyone steps in), but alas, it's Anime Strike.

I_Drive_DSM wrote:
I would assume also that in modern animation this isn't as much of an issue discrepancy as it once was. Animation & coloring is mostly done on computer now. Even if any sort of minor lip or facial discrepancy exists it can easily be rectified in a physical or subsequent release. In the cel days this had to have incurred more work simply due to having to know what frames to say re-use or time at a particular point.

I'm also imagining a scenario where an editor is like "Oh man this lip movement is one frame off. Lemme touch this up right quick and we'll be on our way."


Well, technically, they'd still have to know what frames to change and what to change about it. It's just less time-consuming than it used to be.

HolsteinCrab wrote:
I'm also willing to bet it's the dubbing after animation process that explains why ad-libbing is such a cardinal sin in Japanese voice acting. The Mawaru Penguindrum Japanese commentary had a few anecdotes of seiyuu either accidentally slipping ad-libs into scenes (and apologizing afterwards) or thinking they managed to sneak one in and have it accepted only to be told after the take it had to be redone. Technically I'd say anime is visually a very forgiving style for adlibs because the lip movements are deliberately kept vague but a sudden ad-lib can still muck up the tempo of the scene or just not fit so I can see why some directors are very strict about that. Seiyuu in more comedic roles seem to be allowed more freedom in that regard though as Tetsuya Nomura commented how he let Shigeru Chiba throw in ad-libs as Kefka for Dissidia and Kenichi Suzumura for Demyx in Kingdom Hearts II - however those were games so different deal Laughing

I wonder how long it takes the voice acting students at those training schools to learn how to flip the script pages correctly and get to the mic without making a sound. Sure, learning to act is important but without THAT skill you can't even get through the take Wink


I would assume the scripts would be shown digitally now, and all they have to do is swipe the screen to move to the next page, or, even better, the device could hear what they're saying and automatically move to the next page on its own. This might be out of the budget of some studios though. Maybe.

As big as stand-up comedy is in Japan, between rakugo, manzai, and all that, I would love to hear more ad-libbing done, but it'd only really work on comedy, where you can have more spontaneity and you have much less risk of messing up the story. Well, the voice actors could also be clued in on the story from start to finish. That was what was done with Osmosis Jones, which, according to the commentary track, had literally hundreds of ad-libbed lines (though many of them had to be cut because there was nowhere appropriate to put them). Twice Upon a Time had an extremely loose, empty script mostly made out of scenarios and synopses of the scenes, with the voice cast made up almost entirely of stand-up comedians, albeit with very close direction to make sure the actors remained in character and didn't veer off the story as planned. The movie was then animated according to what the actors had come up with.

And there's Eric's Popeye example below, which I only found out relatively recently was ad-libbed, clearing up decades of me wondering why he'd sometimes mutter something and his mouth would stay shut. Basically, it didn't really matter.

I'd also like to point out one more thought: It's way easier to ad-lib something to an episodic series than a serial series. In an episodic series, the story is completely contained within the episode, and the actor can get clued in much easier on what their character does and what they're like. It's a lot harder for a serial, especially one that hasn't been fully written out yet, as there's a good chance an ad-libbed line could derail some future plot point. Something like Osomatsu is very friendly to ad-libbing. Something like One Piece is not. (Colleen Clinkenbeard, FUNimation voice of Luffy, attempted to come up with something silly, only to keep accidentally creating spoilers, such as joking about Luffy riding a dragon hundreds of episodes before it would actually happen.)
Back to top
View user's profile Send private message
HolsteinCrab



Joined: 07 Dec 2017
Posts: 42
PostPosted: Tue Dec 12, 2017 6:40 am Reply with quote
Quote:
A corollary question:

Since animation before acting is the standard in Japan, are there many Japanese voice actors who are comfortable with performing before the animation? I'll bet they're pretty different experiences that require different kinds of practice and readiness, so I wonder if the majority of Japanese voice actors become accustomed to performing while watching the animatics (or storyboards or however far along they are) and actually get uncomfortable with having to perform without seeing anything.


(I apologize for responding in spite of the corollary nature of your question – I couldn’t resist.) Admittedly my familiarity with the matter is a bit outdated due to being interested in seiyuu who are getting long in the tooth but some of them seem to enjoy working on drama CDs for this reason as it grants them more freedom and at the same time places focus entirely on their voice. I imagine for others it can be the opposite because of the increased pressure/focus. Some seiyuu who sound great in anime are mediocre in drama CDs and some who are not that noticeable in anime can be stunning in drama CDs. (And some seiyuu who are great voice actors make terribly hammy stage actors if some of the stage plays I've watched are anything to go by Laughing)

It's worth noting that seiyuu are also trained to dub foreign media such as western movies and tv dramas (as well as plenty of k-dramas these days) so you get people such as Toshiyuki Morikawa, Maaya Sakamoto, Miyuki Sawashiro, Daisuke Namikawa etc doing a lot of live action dubs on top of their anime/game work. The basic technique appears to be the same in that you still have the seiyuu all in the recording studio quietly stepping in and out during the take as opposed to solo recording with ADR. (There was a video circling the interwebs a while back from some 2014 variety show where they followed Kouichi Yamadera at the dubbing studio for an episode of Mad Men and they point at all the Japanese dubbing quirks with garish arrows.) In live action dubbing there’s a different level of complication in that they have to match their vocal performance to another actor’s performance although how faithful they’re expected to be to the original actor’s voice/acting seems to vary wildly. Sometimes they sound as though they’re nigh-on dead ringers in another language, other times it seems they’re going more with how the producers want the image to ideally sound which is probably why Tom Cruise comes off as a lot more rugged in Japanese.

Quote:
I would assume the scripts would be shown digitally now, and all they have to do is swipe the screen to move to the next page, or, even better, the device could hear what they're saying and automatically move to the next page on its own. This might be out of the budget of some studios though. Maybe.


As far as I know they’re still working on paper scripts – though all the recent productions I’ve seen cast pictures of where they pose with scripts weren’t exactly high end productions so who knows? Laughing
Back to top
View user's profile Send private message
HolsteinCrab



Joined: 07 Dec 2017
Posts: 42
PostPosted: Tue Dec 12, 2017 6:42 am Reply with quote
Quote:

HolsteinCrab wrote:
Ufotable is an exception thanks to the insane budgets they manage to boast


Pretty sure that's a myth -- from what I've read, their deal is that they've got all their staffers under one roof like KyoAni, instead of the constellation of sub-subcontracting that defines most anime.


Ah, I guess that also would help give their animation staff more flexibility to re-do and get it right. Laughing It looked high budget to me because of how fluidly a lot of things moved and the smaller movements characters would make in less dynamic scenes that normally there wouldn’t be time for in a TV animation schedule – but if the staff is permanently based there then I suppose there’s more time to work on animating those little things.


Quote:
This confuses me. Because all the stories and experiences I know are different.

Usually, when recording in Japan for Japanese animation that has not yet premiered, the animation is NOT usually done. You’ll have a set of key frames for the actors to put their performances to, but the animation itself is not done yet.

I hate to refer to my own article, buuuut...

http://www.anime-now.com/entry/2017/02/02/000001

Quote the voice of DBZ’s Goku:

“…I apologize to anyone from Toei in the room, but the footage we’re given to go off of are lines. The pictures aren’t completed,” she said, “We all put our voices to the lines.”

While the voice acting I’ve done has had completely done animation, that’s only because it’s dubbing over the Japanese dialogue with an English script.


Given what was mentioned earlier about older seiyuu complaining about unfinished animation/storyboard reels, it seems that this might well be a quirk resulting from the digital processes now involved in anime. In the late 80s and early 90s when Dragonball was being produced, these digital clean-up and painting processes were not the norm (for example Disney only transitioned from cel to digital in 1990 for its feature films and it took a few years for other feature studios and then TV animation to catch up) so it makes sense that the recording wouldn’t be started until the animation was finished, especially as it was also harder to edit the sound in back then and edit in/out footage to meet the time slot requirements the broadcaster demanded. Now that the technology has advanced to the point that bad frames and takes with slip ups can be easily edited and the computers can fill in the colours and effects so much faster, it is reasonable for a production to begin recording with animatics (which can be anything from storyboard reels to key frames to segments of rough or almost finished line art animation) because now they will be able to edit things to fit. So yeah, it’s understandable that the DBZ cast is taken aback by the change in process – they would be used to working with a finished product and not work-in-progress animatics.

(Granted this is partly speculation on my part – I stopped studying animation a few years back and haven’t worked in Japan so I could be wildly off the mark)
Back to top
View user's profile Send private message
leafy sea dragon



Joined: 27 Oct 2009
Posts: 7163
Location: Another Kingdom
PostPosted: Tue Dec 12, 2017 8:42 pm Reply with quote
HolsteinCrab wrote:
(I apologize for responding in spite of the corollary nature of your question – I couldn’t resist.) Admittedly my familiarity with the matter is a bit outdated due to being interested in seiyuu who are getting long in the tooth but some of them seem to enjoy working on drama CDs for this reason as it grants them more freedom and at the same time places focus entirely on their voice. I imagine for others it can be the opposite because of the increased pressure/focus. Some seiyuu who sound great in anime are mediocre in drama CDs and some who are not that noticeable in anime can be stunning in drama CDs. (And some seiyuu who are great voice actors make terribly hammy stage actors if some of the stage plays I've watched are anything to go by Laughing)


That definitely would've been my best guess, that there is a pretty even mix of people who can act without a guide on a screen without a problem and some who just cannot do it. I'm thinking of western voice actors, some of whom can do both kinds very well, some who have become incredibly accustomed to dubbing, and there are some who seem to rarely or never do any voice acting for anything with pre-made animation, though I can't be sure if it's because they can't bring their A-game looking at a screen or just prefer to perform for western animation.
Back to top
View user's profile Send private message
crosswithyou



Joined: 15 Dec 2007
Posts: 2902
Location: California
PostPosted: Wed Dec 13, 2017 1:07 am Reply with quote
The cast generally records off rough storyboard art, afaik. In the best cases they'll kind of be able to tell what's going on with the scene by looking at the footage, but in the worst case they might be very very rudimentary stick figures or sometimes they might not have any video at all.

The lip flaps are kind of pre-timed and there are bubbles on the screen to indicate when a character is talking. They record to those pre-timed bubbles and the animation is done according to that timing as well, so I assume this is how they get things to sync up. I don't doubt there are times when they might have to edit the audio a bit to fit the animation however.

Check out last week's episode of Osomatsu san (2nd season ep 10) to get an idea of what kind of visuals they get when recording. You can see the timed bubbles for the cast to record to (though in the episode those timings go completely ignored LOL). Pretty sure they cleaned up the visuals for the episode a bit and the actual recording is done to rougher images though.
Back to top
View user's profile Send private message Visit poster's website
leafy sea dragon



Joined: 27 Oct 2009
Posts: 7163
Location: Another Kingdom
PostPosted: Wed Dec 13, 2017 8:51 pm Reply with quote
crosswithyou wrote:
The lip flaps are kind of pre-timed and there are bubbles on the screen to indicate when a character is talking. They record to those pre-timed bubbles and the animation is done according to that timing as well, so I assume this is how they get things to sync up. I don't doubt there are times when they might have to edit the audio a bit to fit the animation however.


Sounds like it'd require more than one take. Do they have any indication of exactly how long the line lasts? I can't imagine anyone being good enough to do it in one take without getting lucky. And here, I thought the lines are scrolling across the screen or something to help them with their timing. It must require a lot of practice from the animators too, to not constantly overshoot or undershoot how long the actors will say each line.

That's why I always wondered why anime uses this system first, since it means they'd theoretically have to constantly correct the animation afterward anyway. It sounds like a lot of guesswork that has to be done mostly right, whereas the voices-first method has no guesswork.
Back to top
View user's profile Send private message
crosswithyou



Joined: 15 Dec 2007
Posts: 2902
Location: California
PostPosted: Wed Dec 13, 2017 10:21 pm Reply with quote
leafy sea dragon wrote:
Sounds like it'd require more than one take. Do they have any indication of exactly how long the line lasts? I can't imagine anyone being good enough to do it in one take without getting lucky. And here, I thought the lines are scrolling across the screen or something to help them with their timing. It must require a lot of practice from the animators too, to not constantly overshoot or undershoot how long the actors will say each line.

The cast generally gets the rehearsal video (the video used during recording) along with the script ahead of time for them to practice and get the timing down. (There are time codes on the video too.) They also do a practice run of the entire episode in the studio before the actual recording. At that time, they can make any adjustments to timings they need to make, and the sound director can give them instructions if what they heard during the rehearsal wasn't what they were looking for. FYI, one 24-minute anime episode is generally scheduled for 5 hours of recording in the studio, though it may not always take the entire allotted time for them to finish.

I'm not as familiar about the animation side of things, but I think it's more the actors fitting their dialogue to the animation rather than the other way around, so in that sense the animators don't really need to be too conscious of how the actors will fit the lines in.

And as a little tidbit, when actors record during or after visual production (as is the case with most anime), it is called "afureko," short for "after recording." With this, the actors are the ones who need to sync up their lines to the visuals. When they are replacing lines such as with dubbings, it is called "atereko." And lastly, when the cast records their lines first and the animation is then fit to the audio recording like how ufotable and the West does things, it is called "puresuko." short for "pre-scoring."
Back to top
View user's profile Send private message Visit poster's website
relyat08



Joined: 20 Mar 2013
Posts: 4125
Location: Northern Virginia
PostPosted: Wed Dec 13, 2017 10:31 pm Reply with quote
I've seen a lot of making-of documentaries that make the sessions pretty easy to understand. I'm sure if I looked for another minute I could find a better one than this, but it'll do the job. As you can see in the reflection behind them, they are watching the animation as it plays. Depending on the level of completeness, it might be nearly done, as in this case, or they could be looking at simple animatics(storyboards, some key animation, etc). Either way, whatever has been done will be played for them in real time, and they'll be looking at the screen to help them with timing and prompts.
Back to top
View user's profile Send private message My Anime My Manga
Display posts from previous:   
Reply to topic    Anime News Network Forum Index -> Site-related -> Talkback All times are GMT - 5 Hours
Goto page Previous  1, 2, 3  Next
Page 2 of 3

 


Powered by phpBB © 2001, 2005 phpBB Group