A bidirectional AI model combining speech acoustics with transcript meaning reached 90.42% accuracy on the ADReSS Alzheimer’s disease benchmark in a small retrospective study.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results