Jump to content


Photo

On lip-synching audio dialogue


  • Please log in to reply
2 replies to this topic

#1 skarstensen

skarstensen

    Wannabe film-maker

  • Members
  • 30 posts
  • Gender:Male
  • Location:San Jose, CA
  • Interests:Music, Movies, Game Development

Posted 29 September 2015 - 07:41 PM

So, I thought I ran into a bug last night but it turned out to be "as intended".  However, it's very easy to think this is a bug and waste time trying to "fix" it, so I thought I'd mention it here.

 

When using the "Say" action on an actor, the dialog that opens offers you a space to type in the dialogue spoken, as well as link it to an audio file.  Any text typed here will show up on the timeline, so it makes for a good reference point.

 

However, if there is text in this box, Moviestorm will use it for the lip-synch and completely ignore any attached audio.

 

For example, if I attach an audio file of a lengthy monologue, then type the first line of said monologue into the text box (what I actually did, and why I'm posting this) just for reference, then the actor will mouth that first line of dialogue, and that first line only.  They will not synch to the audio at all.

 

If you leave the text box blank, however, Moviestorm will synch to the audio file instead.  Best practice, obviously, is to have both text and audio match exactly... but sometimes this is not feasible, so it's good to know the default behavior in such cases.

 

Best,

Steve


Modern Marvels Entertainment on the YouTubes


#2 Ben_S

Ben_S

    Master Director

  • Mods
  • 4811 posts

Posted 05 October 2015 - 12:41 PM

The text, if present, is used as a guide for a lip sync. The lip sync should be better if there is text matching the audio, than if there is audio with no text - the lip sync module can derive mouth shapes that are needed to say written words (in about 7 languages, but I forget which ones).

 

Having text that does not match the audio is likely to lead to different lip sync to what you might expect.

 

I would suggest chopping your audio up into sections that can be added separately, and ideally putting in all the text for each section.


Ben Sanders
Moviestorm Ltd

#3 spenayoung

spenayoung

    Director

  • Members
  • 148 posts

Posted 27 January 2016 - 01:45 PM

For Libertaria, which was a musical, I had to take all audio and chop it up in Logic (a music software program) then feed it into the program. I typed in lyrics for most of them. I made sure that the words were chopped up enough that they program could understand it. Then I would take a few takes at least, with different lip sync settings and then edit all that in post. Seems like a lot of work, and it was (we had 1000+ audio takes for the project), but that's what we had to do.

 

 

 

I did short clips overall for accuracy then mixed it all in Final Cut. 

 

 

 

 


Sabrina Pena Young
Composer/Filmmaker/Sci-fi Junkie
Libertaria: The Virtual Opera

You must unlearn what you have learned.
Visit My Website


  • Please log in to reply


1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users