Pyro Board. Or flammable sound waves and music. Danish Fysikshow demonstrates a 2-D Rubens' tube (wiki, demo).
Towards the end of the 1800s, there were three primary American groups competing to invent technology to record and play back audio. Alexander Graham Bell worked with with Charles Sumner Tainter and Chichester Bell in at their Volta Laboratory in Georgetown, Washington, D.C., while Thomas A. Edison worked from his Menlo Park facilities, and Emile Berliner worked in his independent laboratory in his home. To secure the rights to their inventions, the three groups sent samples of their work to the Smithsonian. These recordings became part of the permanent collections, now consisting of 400 of the earliest audio recordings ever made. But knowledge of their contents was limited to old, short descriptions, as the rubber, beeswax, glass, tin foil and brass recording media are fragile, and playback devices might damage the recordings, if such working devices are even available. That is, until a collaborative project with the Library of Congress and Lawrence Berkeley National Laboratory came together to make 2D and 3D optical scanners, capable of visually recording the patterns marked on discs and cylinders, respectively. [more inside]
What is up with Noises? A fascinating explanation of why we hear sounds and music the way we do. It's a long video, but it's worth it!
What is the relationship between the optical groove in a record or wax cylinder and sound, and how can we use this to recover analog recordings from the past? Dr. Carl Haber explains IRENE (.pdf; begin at slide 44 for audio samples).
University of Arizona physicists have discovered how to turn single molecules into working transistors. The research could result in much smaller, more powerful computers and other devices with the ability to process many more channels of high-resolution audio and video than current products can manage. The abstract is available in PDF.
There's been a lot of talk of late about signal-to-noise ratios here on MeFi (er, Ashcroft who?...). Generally, we think of noise as something that always degrades the quality of a signal. Sometimes, however, the opposite can be the case. Here's a neat little demonstration of a non-linear system in which noise can be used to amplify a signal that would otherwise be too be faint to detect any other way. It exploits a phenomenon known as Stochastic Resonance.