Comment

Comment by sova

Spectograms of each note will also make it strikingly clear that there is a dominant frequency invoked and overtones (harmonics) also being invoked that give the sound its full sound-profile. If sound is atmospheric texture, the overtones are irreplaceable grooves in the ether.

The simple sine wave is exactly one dominant frequency in a spectrogram, a line. Instruments such as a trumpet will have upwards of 12 overtones, parallel lines, lessening in strength.

One interesting idea that came to me last night was trying to reproduce the physical 3D model of an instrument based on its spectrographic fingerprint. With enough samples, this ought be possible, and with a 3D printer one might even be able to create interesting physical instantiations of instruments based on spectrographic fingerprints. One could even create never-before-seen instruments based on a generated spectrogram, in an interesting radar-to-ocean operation (as opposed to ocean-to-radar, how radar normally works). Maybe topography-from-radar is a clearer way to state the same.

Generating audio from spectrograms is an open problem and I would love to see more open-source work in this domain.

parent sova 5 years ago

Discussion

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton