An implanted interface that lets a person with paralysis communicate at close to the speed of natural conversation, reliably, from a device that works for years. This entry maps research; it is not medical advice.
Scope
In: intracortical and other implanted interfaces that decode attempted speech or movement into text, sound or control. Out: non-invasive interfaces, and neural stimulation therapies. The recording electrodes are covered by neurotech/long-lived-neural-electrodes.
Communication rate headline0.4 orders of magnitude to go
Words per minute a person communicates through an interface, with the vocabulary size and the word error rate stated in the conditions; across modalities the field also reports the information transfer rate in bits per minute. Higher is better.
Current (2023-01-01)62 words min⁻¹
Target160 words min⁻¹
Limit–
Conditions. Speech-to-text decoding of attempted speech, intracortical arrays, one participant with ALS, large vocabulary.
Why this target. The speed of natural conversation as stated in the source paper, the rate at which a spoken exchange flows without the user waiting.
Note. as_of is the publication year; a later study in the same field reports about 32 words per minute in self-paced conversation (Card et al., NEJM 2024), so the figure depends on the task. One participant.
The 2023 demonstration decoded 62 words per minute, which the authors describe as beginning to approach the speed of natural conversation, 160 words per minute. The result comes from one participant, so how it generalizes across people is not established.
Long-lived neural electrodesDecoding depends on recordings from implanted electrodes that must keep working for years.
Energy-efficient AI inferenceDecoding runs a neural network on the neural signal; running it in a wearable or implanted device needs low energy per inference.
Required by
Nothing in the atlas depends on it yet.
Arrows point from a technology to what it requires. Select a node to open it.