Brain-to-Speech Decoder (Conv1D + BiGRU + HiFi-GAN, trained on VOCALMIND sEEG) — Math in LLMs Research Map

MATH · IN · MODELS

Brain-to-Speech Decoder (Conv1D + BiGRU + HiFi-GAN, trained on VOCALMIND sEEG)

University of Maryland College Park (Maghsoudi & Mishra)

Structures found in this family (1)

Linear Subspace

By model (1)

Custom Conv1D-encoder + 3-layer BiGRU + HiFi-GAN brain-to-speech decoder (trained on real human sEEG, VOCALMIND)

Linear Subspace

Observations (1)

Causal patching localizes vocalized/mimed/imagined speech transfer to a compact subspace in a real brain-to-speech decoder

Papers

Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes (2026)