Integrating Language Models into Listened and Imagined Speech Decoding from MEG
It is found that imagined-speech decoding benefits more from the language model than listened-speech decoding, and the best-performing balance between neural and language-model evidence shifts toward the language model, and the gain over neural-only decoding is larger.
Maryam Maghsoudi, Sai Samrat Kankanala, S. Shamma et al.
· 0 citations