From a technical point of view, I already know how to add sounds to animations. I add sound strips in the Video Sequence editor for ambient noises, and use Speakers when I need 3D attenuated audio.
Still, I don’t think I understand all the ins and outs of properly doing sound in Blender yet. Most of the time I just get appropriate sound files and try to match them to keyframes where collisions take place, or loop them on noise sources in the world. In many cases that’s all the animator needs to do for a correct effect, but sound can sometimes go into greater levels of detail too.
For example, imagine a plastic ball bouncing around a wooden table. You want the correct “pong” effect whenever the ball hits a surface, which can include pitch changing based on the velocity of the ball. Or a wooden crate rolling around, each corner generating the proper noise based on the mass of the crate that hits the floor that moment. Or think of a cement wall being smashed to pieces and little pebbles of dirt raining to the ground… how do you realistically do the audio for that?
I’m wondering if anyone made a more in depth tutorial about sounds in Blender animations, primarily using Speakers with stereo direction and attenuation. NOT a technical guide on how to play a sound at a certain frame… but how to properly think and match sounds, and what tricks can be used to make every detail sound as correct as possible.
Well, in essence yes, that’s mostly it. My curiosity was if there other details you generally need to consider, or other features in Blender, for sound to be realistic in very exact situations.
It is called Foley and is an art all to itself. Just as modeling, lighting, camera work, animation are their own disciplines of the same craft.
If all you have is Blender then by all means proceed, it can be done. But for my personal sound effects work I have been using Adobe Audition quite a bit. Before that I was using, the now defunct, Sound Track Pro. Adding sound to video after the video has been rendered is so much more easier than laboring over endless previews one-by-one. So I would suggest render all your video and then add the sound in a true sound editing program.
I found this example BLEND file a while back. It shows how to render out audio with video from Blender and there is a small example python script in the text window that shows how to load a wave file into an exact frame #. So if you make a list of sound effects you could cobble together a script to load all the sound effects at the desired time.
@Atom: Nice, that is some useful info indeed! I wasn’t aware of Foley, and the concept described there sounds like a good way. I was already doing something similar: If I had to animate a metal object hitting wood for example, I would start recording audio, throw my keys on the table several times, choose the best sound from all attempts, then save it as wav and place it at the correct frame on the timeline. Watching the animation play and making sounds in real time feels like an even better way, even if it probably takes more practice.
As for software, I used Adobe Audition in the past. But now I’m on Linux and giving up on closed-source software, so I use Audacious for sound. Much more limited but it’s pretty good for anything I might need.
I just remembered about old time radio sound effects. They were not recorded but played live. It’s all in the technique! Here is good example of sound effect method: