Animating Fire with Sound

ACM SIGGRAPH 2011

We propose a practical method for synthesizing plausible fire sounds that are synchronized with physically based fire animations. To enable synthesis of combustion sounds without incurring the cost of time-stepping fluid simulations at audio rates, we decompose our synthesis procedure into two components.

First, a low-frequency flame sound is synthesized using a physically based combustion sound model driven with data from a visual flame simulation run at a relatively low temporal sampling rate.

Second, we propose two bandwidth extension methods for synthesizing additional high-frequency flame sound content: (1) spectral bandwidth extension which synthesizes higher-frequency noise matching combustion sound spectra from theory and experiment; and (2) data-driven texture synthesis to synthesize high-frequency content based on input flame sound recordings. Various examples and comparisons are presented demonstrating plausible flame sounds, from small candle flames to large flame jets.

Paper | Citation | Video | Video (Youtube) | Press | Source Code | Presentation Slides | Acknowledgements

Paper:

FireSound2011.pdf (PDF, 11MB)

Citation:

Jeffrey N. Chadwick and Doug L. James, Animating Fire with Sound, ACM Transactions on Graphics (SIGGRAPH 2011), 30(4), August, 2011 (TeX)

Video:

Paper video: FireSound2011.mov (Quicktime, H264 59MB)

Animation only (no sound)
Low frequency result
Bandwidth extension result (alpha = 2.5)
Bandwidth extension result (alpha = 3.0)
Bandwidth extension result (alpha = 3.5)
Texture synthesis result

Animation only (no sound)
Low frequency result
Bandwidth extension result (alpha = 2.5)
Bandwidth extension result (alpha = 3.0)
Bandwidth extension result (alpha = 3.2)
Bandwidth extension result (alpha = 3.7)
Texture synthesis result

Animation only (no sound): v1 | v2
Low frequency result: v1 | v2
Texture synthesis result (gas flame training data): v1 | v2
Texture synthesis result (burning oil can training data): v1 | v2

Animation only (no sound)
Low frequency result
Bandwidth extension result (alpha = 2.5)
Bandwidth extension result (alpha = 3.0)
Texture synthesis result

Animation only (no sound)
Low frequency result
Bandwidth extension result (alpha = 3.0)
Texture synthesis result (torch training data 1)
Texture synthesis result (torch training data 2)
Texture synthesis result (burning oil can training data)

Press:

New Scientist - Fire-breathing dragon roars to life

Source Code:

Source code for the spectral bandwidth extension and sound texture synthesis algorithms used in the paper can be downloaded here. The bandwidth extension algorithm is implemented in Matlab, and C++ code is provided for sound texture synthesis.

Sound texture synthesis: tar.gz | ZIP
Bandwidth extension: tar.gz | ZIP

See the README documents in each archive for instructions on how to compile and run our software.

Training inputs for the sound texture synthesis algorithm are selected from the Ultimate Fire sound library. Permission to provide a few of these audio clips as examples in our source code package has been generously granted by The Recordist.

Compilation of the sound texture synthesis routines requires the ANN nearest neighbor search library.

SIGGRAPH 2011 Presentation:

Keynote slides (zipped, 247MB)

Acknowledgements:

The National Science Foundation (HCC-0905506)
The Natural Science and Engineering Research Council of Canada
The Alfred P. Sloan Foundation
The John Simon Guggenheim Memorial Foundation
Intel (ISTC-VC)
Pixar
Autodesk
Vision Research
Side Effects Software (for Houdini 3D animation tools and Mantra renderer)

Copyright Notice:

The documents contained in these directories are included by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.