The paper investigates whether large models are necessary for speech brain‑computer interfaces (BCIs) and introduces BEST, a two‑stage framework that uses hierarchical intermediate representations and a compact decoder to map neural activity directly to text. BEST achieves low word error rates (6.19% and 1.62%) on the Brain‑to‑Text ’24 and ’25 benchmarks while using only 3.85% of the parameters of comparable methods. The study also highlights that decoder scaling yields non‑monotonic gains, cross‑modal alignment is more critical than model size, and overfitting harms sequence‑level performance.
By Zehui Feng, Weichuan Wang, Xiaohan Chen, Cuntai Guan, Ting Han
Brain2Qwerty v2 is a model that decodes natural sentences from real‑time magnetoencephalography (MEG) recordings, achieving an average word error rate of 39% across 22,000 sentences typed by nine subjects. The model uses character, word, and sentence‑level representations and shows that decoding accuracy improves log‑linearly with more data, narrowing the gap to intracranial brain‑computer interfaces. AI contributes by replacing hand‑crafted event detection with deep learning, fine‑tuning large language models for semantic extraction, and employing AI agents to refine the decoding pipeline through automated code development.
By Mingfang Zhang, Jarod L\'evy, Cedric Rommel, J\'er\'emy Rapin, Corentin Bel, Julie Bonnaire, Daniel Nieto, Pierre Bourdillon, Svetlana Pinet, St\'ephane d'Ascoli, Thomas Moreau, Jean-R\'emi King
SCRIBE is a framework that uses a neuroscience-informed semantic space and an ensemble of large language models to generate interpretable, word-level semantic representations of arbitrary text. In a behavioral study with 921 participants, SCRIBE’s representations varied systematically with context and captured 75‑81% of the estimated human noise ceiling. When applied to brain activity recorded while participants listened to naturalistic stories, SCRIBE predicted neural responses in language-selective brain regions better than non-contextual representations and explained unique brain-predictive information beyond syntactic features.
By Camila Kolling, Omer Moussa, Emin \c{C}elik, Mariya Toneva
LibriBrain100 is a new large‑scale MEG dataset for speech decoding that contains over 100 hours of high‑quality recordings while subjects listened to naturalistic continuous speech. The dataset more than doubles the size of the original LibriBrain release, with a record 80 hours from a single subject and additional 40‑minute recordings from 32 subjects. The authors demonstrate the value of deep within‑subject data and broad multi‑subject data by achieving state‑of‑the‑art word‑classification performance and showing that supervised fine‑tuning can compensate for limited per‑subject data, all supported by open‑source tools and a public competition leaderboard.
By Francesco Mantegna, Dulhan Jayalath, Gereon Elvers, Tasha Kim, Benjamin Ballyk, Alex Fung, SungJun Cho, Teyun Kwon, Luisa Kurth, Miran \"Ozdogan, Gilad Landau, Pratik Somaiya, Natalie Voets, Mark Woolrich, Oiwi Parker Jones
arXiv:2607. 19394v1 Announce Type: cross Abstract: Generalizing across subjects remains challenging in invasive neural recordings because electrode configurations, anatomical structures, and neural signal patterns vary substantially across individuals.
By Ji-Hoon Heo, Aleksandra Joanna Wisniewska, Seo-Hyun Lee, Seong-Whan Lee
The paper introduces MD‑SigLIP, a margin‑regularized structured semantic alignment framework that directly aligns brain embeddings with text embeddings in a shared semantic space for retrieval‑based decoding. It builds on duplicate‑aware sigmoid contrastive learning and adds a listwise margin‑regularized term to enforce structured ranking constraints between positive semantic clusters and negative samples. Experiments show that this approach achieves state‑of‑the‑art retrieval performance in both full‑vocabulary and subset evaluation settings.
By Jiaqi Wang, Huawen Hu, Shu Zhang
With the rapid advancement of large language models, brain-language decoding has achieved remarkable progress. However, it remains unclear whether decoded content genuinely reflects neural representat...
arXiv:2610.23745v1 Announce Type: cross
Abstract: Recent large audio language models (LALMs) have extended their capabilities beyond lexical speech understanding to tasks requiring paralinguistic and...
By Koki Ryu, Hitomi Yanaka
arXiv:2608. 08569v1 Announce Type: new Abstract: Recent advancements in Speech Large Language Models have demonstrated remarkable capabilities in understanding complex audio tasks.
By Wenxu Jia, Dongjie Fu, Xize Cheng, Fangming Feng, Linjun Li, Wenshi Chen, Yingming Li, Zhou Zhao, Tao Jin
arXiv:2605.27840v2 Announce Type: replace-cross
Abstract: Audio tokenizers are fundamental to unifying audio understanding and generation. Understanding requires high-level semantics, while generatio...
By Zhisheng Zhang, Xiang Li, Yixuan Zhou, Jing Peng, Guoyang Zeng, Zhiyong Wu
The paper introduces Local Prototype Reconstruction (LPR), a lightweight regularizer that encourages bridge embeddings in speech‑to‑LLM systems to stay close to the frozen LLM’s input‑embedding neighbourhoods. By measuring lexical manifold compatibility with a fixed, head‑free, timestamp‑free diagnostic, the authors show that next‑word prediction and sentence‑level contrastive pretraining do not fully capture token‑level lexical compatibility. Experiments on multilingual ASR and speech translation demonstrate that LPR improves transfer, especially for translation and low‑resource adaptation, and that the diagnostic correlates with downstream gains across objectives.
By Xinnian Zhao, Chia-Hua Wu, Pu Wang, Hugo Van Hamme
The paper introduces a context‑aware neural grapheme‑to‑phoneme (G2P) system for unsegmented languages like Japanese, using a discriminative conditional random field over a word lattice built from dictionaries. It addresses data scarcity by generating over two million sentences with large language models. Experiments show the method surpasses traditional morphological analyzers and neural sequence models, achieving 99.62% target word reading accuracy and very low phoneme error rates on the Joyo‑Kanji‑Yomi benchmark.
By Rui Hu, Zhenpeng Zhan, Xiaolong Lin