The paper introduces Open‑Vocabulary Mutual Information (OVMI), an information‑theoretic metric that quantifies how much of a user’s intended speech a speech brain‑computer interface (BCI) can convey relative to a reference word distribution. OVMI enables comparison of systems that use different vocabularies, recording methods, and datasets, revealing that conventional metrics like accuracy and word error rate can overstate performance. Using OVMI, the authors compare existing speech BCI systems, expose trade‑offs between vocabulary coverage and decoding accuracy, and show that optimizing vocabulary selection for OVMI can improve accuracy by up to 16.3% across three speech domains.
By Dulhan Jayalath, Benjamin Ballyk, Oiwi Parker Jones
The article surveys brain‑to‑language decoding, covering tasks, neural signals, methods, and evaluation across invasive and non‑invasive modalities. It traces the field’s evolution from constrained recognition to text generation, streaming speech, and facial animation, linking tasks to neural populations and decoder representations. The review highlights complementary decoding targets, shared representations, and the growing importance of calibration, feedback, and user control for online communication, while proposing a five‑level future trajectory toward bidirectional cognitive exchange.
By Yiqian Yang, Yiqun Duan, Chenyu Liu, Yiqi Wang, Xinliang Zhou, Chin-Teng Lin, Yu Zhang
arXiv:2609.27650v4 Announce Type: replace
Abstract: Brain-to-language decoding translates neural activity associated with language production, internal speech and perception into linguistic or expres...
By Yiqian Yang, Yiqun Duan, Chenyu Liu, Yiqi Wang, Xinliang Zhou, Chin-Teng Lin, Yu Zhang
arXiv:2609.36736v1 Announce Type: cross
Abstract: Speech brain-computer interfaces (BCIs) aim to restore communication by transforming neural activity related to speech, language, or communicative in...
By Moein Khajehnejad, Forough Habibollahi, Tommaso Boccato, Margarida Sousa, Michal Olak, Francesco Jamal Sheiban, Matteo Ferrante
arXiv:2607. 18749v1 Announce Type: new Abstract: Translating brain signals into text could restore communication for people with severe paralysis, yet practically usable systems to date rely on invasive electrocorticography (ECoG).
By Zihan Zhang (Research Center for Social Computing and Interactive Robotics, Harbin Institute of Technology), Yu Bao (Research Center for Social Computing and Interactive Robotics, Harbin Institute of Technology, Shanghai Innovation Institute), Xiao Ding (Research Center for Social Computing and Interactive Robotics, Harbin Institute of Technology), Tianyi Jiang (State Key Laboratory for Novel Software Technology, Nanjing University), Kai Xiong (Zhongguancun Laboratory)
EEG-Captioner introduces a new EEG-to-language framework that grounds language generation in measurable EEG signal properties. The authors built a large dataset of 160,781 EEG segments annotated with question‑answer pairs targeting spectral, spatial, temporal, and lateral properties, and paired each segment with a feature‑grounded caption. Their Dual‑Domain EEG Encoder jointly models temporal waveforms and spectral‑spatial representations, achieving a macro‑averaged balanced accuracy of 88.81% on signal‑level QA and improving caption factual consistency from 82.06% to 83.46% compared to caption‑only models.
By Leyuan Jin, Jiahao Mei, Yixuan Yuan, Shimeng Li, Mengyue Wu, Yue Ding
The paper introduces Language as Supervision (LaS), a method that aligns neural activity with natural‑language class descriptions via learned projections and uses candidate text prototypes for classification. Evaluated across seven encoder architectures on visual, motor, and speech decoding tasks from scalp and intracranial recordings, LaS boosts speech‑decoding accuracy by up to 17.15 percentage points compared to traditional class‑label supervision. The approach also enables reusable candidate sets, subject‑ and session‑adaptation, and conditional unseen‑frequency decoding in steady‑state visual evoked potentials, with dual‑sided attribution visualizations highlighting both neural and textual contributions to class matching.
By Xuan Wen, Yonghao Song, Yuhong Zhang, Guoliang Li, Tzyy-Ping Jung, Xiaorong Gao
arXiv:2606. 26880v1 Announce Type: cross Abstract: Language-model representations provide structured, high-dimensional annotations of naturalistic language stimuli and can serve as informative neural predictors during comprehension.
By Xiao Jia
The study evaluated whether imagined speech decoded from scalp EEG can transfer across individuals using three public datasets. Across 482 participant pairs, cross‑subject decoding accuracy was essentially at chance, while within‑subject decoding only surpassed chance in one dataset. The results suggest that imagined‑speech content does not generalize between people and that larger single‑condition datasets are more beneficial than further model development.
By Blessing Isoyiza Adeika, Onyema Osuagwu
SHINE is a Sequential Hierarchical Integration Network designed to reconstruct speech envelope and Mel spectrogram from EEG and MEG recordings. It uses a residual sensor adapter, dilated-block states for temporal depth, and a target- and time-dependent gate to fuse hierarchical and attention-enhanced context predictions. Across two EEG and two MEG datasets, SHINE achieved the highest mean envelope and mean-Mel Pearson correlations among nine baseline methods and ranked second in the NeurIPS 2025 PNPL Competition’s speech-detection Extended Track.
By Xiran Xu, Yujie Yan, Songyi Li, Linze Zheng, Zifeng Zhang, Mochu Dong, Jing Chen
arXiv:2608. 20186v1 Announce Type: new Abstract: Non-invasive decoding of inner speech faces a fundamental data problem: a corpus pairing brain activity with a person's spontaneous inner monologue cannot be collected, and the available proxy paradigms (cued repetitive and retrospectively reported generative inner speech) are slow to acquire, poorly time-locked, and subject compliance is unverifiable.
By Ingo Marquardt, Anthilia Alchanat, Priyanka Jain
arXiv:2607. 03925v1 Announce Type: new Abstract: EEG foundation models have shown strong potential in learning generalized representations across subjects and tasks.
By Weibin Li, Wendu Li, Yushan You, Chen Wei, Quanying Liu