https://en.m.wikipedia.org/wiki/Soundex
Maybe someday AI can generate novel media encodings. The rest of the blanks are filled in via inference client-side.