Loading
What changed between revision 1 and revision 2, compared over the article's markdown source with its frontmatter — so a change to status, evidence or reversibility shows up as the edit it is.
Older · revision 1
Import· initial corpus import, not an edit
Initial import of content/brain-computer-interface.md — the filesystem corpus, unchanged. Not an edit.
19.1 kB
Revision 1
Revision 2
lines 67–73 → 67–7366 unchanged lines not shown
*Decoding* maps features to intent. Early systems used linear filters and the population-vector*Decoding* maps features to intent. Early systems used linear filters and the population-vectorapproach inherited from primate motor physiology; current systems use Kalman filters, recurrentapproach inherited from primate motor physiology; current systems use Kalman filters, recurrentneural networks, and increasingly transformer-style sequence models paired with language models.neural networks, and increasingly transformer-style sequence models paired with [language models](https://aiwiki.ai/wiki/large_language_model).[[neural-decoding]] is where most of the recent performance gains have come from — the electrodes[[neural-decoding]] is where most of the recent performance gains have come from — the electrodesin the best 2025 speech systems are not fundamentally better than those used in 2012.in the best 2025 speech systems are not fundamentally better than those used in 2012. removed, struck through added, underlinedLine numbers count the serialised markdown of each revision, frontmatter included.
282 lines are the same in both revisions, and 276 of them are not shown. Neither revision here is necessarily the current article — read Brain–computer interface for the text this wiki stands behind today.