Is it possible to create 3d sound in blender? For example to animate bee flying around the camera.
May be someone know some python scripts to do it.
you are right this is something important, i saw once a guy that made puredata integration trough python, and that could be made i think trough it, but it is something very involved i think, not sure.
Actually itās very simple to create bee effect and I even have an idea how to create voice like effect, but now Iāve got no idea about creatinā echo effect or decreasing volume after sound go through an objectā¦:spin:
your method can be very simple, yet different from what im thinking about, care to share with us your ideas?.
There was a discussion in late '06 about this. I think there is a growing support for a full sound suite for Blender, with things like microphones, sound objects (sources that make sound), sound deflection (for very cool echo-like effects), dampers, etc. 3D MAX has the Foley plug-in, which does all of this, so there should be nothing to stop a similar package from being made for Blender. Except that peopple are probably more trained in making graphical stuff than audio. But if there is someone out there with some hard OpenAL skills, I would definitely love to see this become a part of Blender, as well.
And I have a bunch of other sound features that I would love to see included, but I will not bore you with them here.
The method is animate amplitude for wav file according to distance betwen two empty object - ear (static) and bee(child of bee model). You actually have to make it twice for left and rigt ear, becouse blender support only one channel AFAIK
Some one who is expert in C can help. Take the Audacity source code and integrate with Blender. Audacity is also free like Blender.
Audacity has lot of features like Sound generation, change pitch, change speed etc.
I think audacity is useless for 3D sound :no:
I do not know Audacity in depth, but from what I have seen, donāt be so sure! It is mainly a matter of integrating a piece of basic sound editing with a fictional environment based on the 3D information stored in a blender file. It will take some work, yes, but what is needed is for a new software component to look at where objects marked with āsound parametersā (source, capture, dampen, reflect, to mention the four basics I think would be all that is needed for a first version) and use something like the Audacity code to manipulate sound accordingly. If you know how to make the software render a few spots of VISUAL 3D (like a few dots in a rendering engine), you can produce something that renders AUDIO 3D āspotsā. Audacity seems to be for sound what the material editor in Blender is for visuals.
Of course, my programming skills are way too low to go into details, but with a basic OpenSource sound suite like Audacity at hand, I am very certain that BlenderSound is not that far away. We just need someone who can handle both camps of codeā¦ THAT might be a bigger problem!
i guess that audacity is good for audio treatment, but i think that blender is a generator tool, like it generates things, images for instance, but this could be extended to the generation of sounds, and one nice āengineā to keep in the graphics talking is puredata, it allows you to generate sounds and do also the audio treatment, im not yet proficient in C to give it a try but i will when i learn.
I think you donāt exactly understand.
to generate a sound- in the purest form of the word - means to mathematically generate the waveform. That means youād have to tune tons of parameters to get say a bee buzz.
what would work better, and what I believe the others are talking about, is to have the basic sound of an object through its lifetime in a file, (i.e. a bee buzz that lasts the length of the animation), then, a script would get a vector between ear and bee that would adjust the stereo of the audio track (to make it sound like itās on the left or right), and the volume. also needed would be to get a distance vector between ear and bee that would be used to edit pitch for approaching and go away from the ear.
code from audacity would work well, it already has pitch variance support and stereo and volume support, though it would be better to use openAL IMHO.
Jessethemid: Yes, that is what I at least was thinking. Add to the features that objects hit by sound may dampen, block, distort or reflect (echo) it, and you have one very powerful moviemaking component for Blender! Whether Audacity is a shortcut or not is something I do not feel qualified to say. It just looks like it might be a help in a first version (mangle stuff to fit better in a second version).
I have myself thought about a feature that actually synthesizes sound from object data (two cars collide, how do their component objects react and interact to create sound, etc.), but that would DEFINITELY be a later addition; it is not the main concern of a proper Blender Sound Suite.
http://www.blender.org/documentation/intranet/docs/develop/audio/index.html
Hi i was researching blender sound system and came across that doc, acordingly to it blender already has 3d sound support only in game engine, if somehow we could use samples with more quality than is supported and bake a animation sound we would have a perfect system for animations also, something like is done with rigid body physics bake to ipo but with 3d sound it even has the doppler effect.
VERY interestingā¦ now we just need to figure out how to bake this sucker into an animation :eek:
yeah it would need some code.
5.1 plz (6 channel). Any meaningful animation (ED2) needs to have 5.1. I would only need to assign a speaker to an object and tie it to an input sound track, and parent the 5 microphones with the camera, almost always in a fixed setting, with a pickup setting on the microphone (front 3 gain more than rear), and material setting slider for aural reflectivity. the .1 is the bass and can be extracted from all channels through a low-pass filter. In ED, for example, there would be a speaker for Emo, a spkr for Proog, and two or so speakers for ambient sounds (the machinery). just my 2c worth. Sound travels essentially the same as light, so a lot of rendering math should be able to be reused (a speaker is an area lamp that emits sound)
This would allow me to ārecordā 2 channel, 2.1 channel, 3 channel, 3.1, and 5.1 effortlessly.
I dont think making the sound is so important; almost all pros use a recorded foley track, even for like bees. What is important is getting the balance and reflectivity right to match the scene, and if Blender could help with that, that would be awesome. Even if it was just putting out a fader list that showed the volume levels of each audio source over time.
Wrong. You could make the sound of a bee easily using the fade in and fade out effects.
Quote:
Originally Posted by PranksteR
I think audacity is useless for 3D sound :no:Wrong. You could make the sound of a bee easily using the fade in and fade out effects.
i think he was referring to a audacity integration and a automated 3d sound system.
5.1 plz (6 channel)ā¦
Hmm. Something I would like to do with my animations. Iāve been working on a way to achieve this. I tried to do it with CSound (but it crashed whenever it got longer than about 117? frames). Since then, I have got a CVS copy (I last updated it around 16th? February 2007) and added a system for 3D sound. With the right camera setup (I used cameras for microphones), 5 channel sound can easily be achieved. I donāt know how true to science it is, but the results seem quite promising. I might release a patch someday, or see if the developers of Blender would like to include it in the BF tree. Does anyone have any opinions on this?
I think this āBee effectā your looking for is called the āDoppler Effectā :- http://en.wikipedia.org/wiki/Doppler_effect
For this, 5.1 is not required. Just the same as you can see in 3d with 2 eyes, you can hear in 3d with 2 ears. The minor problem being that it is difficult to place your head (as an audio camera if you like) in the 3d space without headphones.
Besides that, It does seem like Audio is a space where Blender could improve.
Small thoughts, or a starter for 10ā¦ maybe looped wav files could be attached as properties to objects, and 2 microphones could be placed either side of the camera. As you get nearer objects or move around them, simple trig can deduce the volume of that wav in each mic and render that to a soundtrack wav file?
It would take some attacking in the form of codeing but this seems to be the best line of attack to me!
J