Observed arrival · 2026-11-17
Ileanor, an artificial organism built around sound
A French solo project lays out a persistent neural network meant to learn audio patterns without pretraining.
- For
- Readers exploring experimental neural networks and auditory perception
- Worth noticing
- The project says every proposed mechanism must come with a measurement showing that it exists.
Field notes
The page organizes the project around an audio pathway: an artificial cochlea divides sound into frequency bands; recurring rhythms and motifs are meant to form representations, with sequence order held in internal time. It says sleep would replay and reinforce some structures while allowing others to fade, and that memory should persist across sessions. Voice and vision appear as future directions; the extract offers the design account, not measured results or a demonstrated interface.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue