exploits

The VisualAudio project was proposed by Stefano Cavaglieri of the Swiss National Record library in Lugano. The idea is based on 2 observations:

On a phonographic disc, the information is extracted via a needle that follows the side position of the furrow. If we observe this furrow by means of a microscope, we can see its radial curve. That means that the musical information is visible.

Many old discs, particularly the originals of the radiophonic productions, unique versions, are terribly damaged: this excludes any reading by a mechanics way, preferring a reading without contact.

4 steps:

The first one is the most critical: photograph the vinyl. Just once at the beginning of the project, then you can take the disc away. And as we would do for a good listening, it's necessary to clean it before, to withdraw dust and other foreign particle. Then the result is now on photographic film.

The second step : scan the picture. It is carried out by a device especially created for this purpose: it turns around the picture and scan it part after part. The result looks like this:

The third step is treating the image: correct the imperfections, determine the position and movings of the furrow. They also make other corrections thanks to physical knowledge thay have of a vinyl record. For example, the interpolation of the furrow in case it's cut.

And the last step: extracting the sound by means of computer, thanks to picture's information. This signal is treated one more time by filters band-pass in order to restore only the band-width of the master. Unfortunately, the website provides too few details on this step which I find the most interesting one.

Nevertheless, here is an example of the result:

And it's true, above the traditional crackings of a vinyl record, the noise is still important. The final objective of this project is to get the best signal-to-noise ratio. According to the article, the quality of a disc 33t will never be reached.

During the tests on the first prototype, they measured signal-to-noise ratios from 15 to 20 dB. This is explained by variations due to lighting (50Hz), vibrations of the rotary engine, and the presence of dust when the picture was shot.
Improvements are continuously made in order to eliminate these problems. The development of new algorithms to detect the furrow (subpixel estimate, filtering, modeling of the furrow,...) should also allow a significant improvment of the quality.

More simply, the ELP Laser Turntable is just a laser turntable for vinyl records. It reads the vinyls without touching them, with a laser, in order to keep it intact.

Let's conclude by this strange story, the story of Doctor Arthur Lintgen, a physicist of Philadelphia. Physicist and audiophile apparently, since he's able, people say, to recognize a track of classical music only by looking at the vinyl record! He would do this exploit by examining the space and the contour of the furrows. From this examination, he can determine the number of movements, as well as the volume and the use of the drums. By searching in his knowledge about classical music, Lintgen can identify precisely which track matchs with his observations. People say.

Source : the website of EIA-FR (no date)

For those who do not know : Signal-to-noise ratio (SNR) is an engineering term for the power ratio between a signal (meaningful information) and the background noise. Because many signals have a very wide dynamic range, SNRs are usually expressed in terms of the logarithmic decibel scale. In decibels, the SNR is 20 times the base-10 logarithm of the amplitude ratio, or 10 times the logarithm of the power ratio.

Labels: exploits