Peer-reviewed research
A high-performance speech neuroprosthesis
Francis R. Willett et al., Nature, 2023 · DOI 10.1038/s41586-023-06377-x
Read the originalRead on October 16, 2026
- What it is
- A Stanford-led study in which a woman who could no longer speak because of ALS had a brain implant that decoded her attempted speech into text on a large vocabulary.
- How much weight it carries
- Peer-reviewed in a leading journal. One participant, a surgically implanted device and a research setting, so it shows what is possible, not what is available.
Facts taken from this source
- A brain-computer interface decoded one participant's attempted speech into text at 62 words per minute, with a 23.8 percent word error rate on a 125,000-word vocabulary.
“Our participant’s attempted speech was decoded at 62 words per minute, which is 3.4 times as fast as the previous record” (Abstract)