By the way, and for the avoidance of doubt/suspicion: none of the "audio etymology" sequences that I've created and am posting here involve any use of AI, LLM's, neural nets, none of that. All are made from original audio recordings of real people speaking words in their various languages, which are then subjected to good old-fashioned signal processing (speech coding and synthesis). The continua of change are made using straightforward linear interpolation over the space of cepstrograms, which are 2 dimensional matrices encoding the acoustics of the original recordings. Using scripts that I wrote, by hand. #acoustics #phonetics #speech #synthesis #audio #maths #linear_algebra