This piece combines improvisational playing on cello, synth, and drums, with three main technological components. The first component is a purpose-built synthesis/sequencer program*. The piece opens with this program layering a base motif 64 times with a random time offset, creating a blurred, textural reference to the original motif that varies with each performance. The second component is a generative Markov model of phoneme sequences derived from Wikipedia and a collection of scientific papers*. We use the model to generate novel, incoherent speech sounds. The third component is a sentiment-aware model of statements of preference derived from peoples’ actual statements of preference on the internet*. We use the model to generate positive/negative sentiment couplets, recited in synthesized speech.
Статистика страницы на pesni.guru ▼
Просмотров сегодня: 3