eme
145 posts

eme
@emeisazam
sound researcher, composer building @sonic_field

AKOÚŌ is a listening harness for agentic workflows ✿ v0.6 is here with: - /remember command for Earworm integration + sound-memory lineage - portable presets for combining the 14 listening modes - evidence-aware claims, machine-readable system contract and structured metadata

Aleph’s silent speech prototype uses under-chin ultrasound to read tongue movement and predict words without an audible voice. An early glimpse of voice interfaces where the computer is watching articulation, not listening to sound. By @alephneuro 🐚 sonicfield.org/silent-speech-…

This is me talking to my computer without making a sound. After just a month of collecting data, our model is already approaching dictation in accuracy. We were surprised to see that it generalizes to unseen participants as well! (1/n)





sonicLAB has released SSNN, Sinan Bokesoy’s new spiking neural network instrument for synthesis and resynthesis, with a concept developed with @daitomanabe 🐚 sonicfield.org/sonic-lab-ssnn…







