analiza aiqbench
SHINE introduces a novel neural architecture designed to reconstruct speech envelopes and spectrograms from EEG and MEG signals, addressing a core challenge in cognitive neuroscience through advanced sequence modeling. By outperforming nine baseline implementations across multiple datasets and securing second place in the NeurIPS 2025 PNPL Competition, this work demonstrates significant progress in bridging non-invasive brain imaging with high-fidelity audio representation.
SHINE: Sequential Hierarchical Integration Network for EEG and MEG
arXiv:2602.23960v2 Announce Type: replace-cross Abstract: How natural speech is represented in the brain constitutes a major challenge for cognitive neuroscience. Reconstructing the speech envelope and Mel spectrogram from EEG and MEG provides a time-resolved way to study its temporal and spectral structure. Speech-related neural activity spans sensors and temporal scales; extracting these representations while adapting the use of context to each acoustic target is a central problem in speech reconstruction. We propose SHINE, a Sequential Hierarchical Integration Network for EEG and MEG. A residual sensor adapter unifies input dimensions, intermediate dilated-block states retain temporal depth, and a target- and time-dependent gate fuses local hierarchical and attention-enhanced context predictions. Across two EEG and two MEG datasets, SHINE has the highest mean envelope and mean-Mel Pearson correlations among nine local baseline implementations on all eight dataset-metric combinations. SHINE also placed second in the speech-detection Extended Track of the NeurIPS 2025 PNPL Competition. Code will be released at https://github.com/xuxiran/SHINE.
Abstract:How natural speech is represented in the brain constitutes a major challenge for cognitive neuroscience. Reconstructing the speech envelope and Mel spectrogram from EEG and MEG provides a time-resolved way to study its temporal and spectral structure. Speech-related neural activity spans sensors and temporal scales; extracting these representations while adapting the use of context to each acoustic target is a central problem in speech reconstruction. We propose SHINE, a Sequential Hierarchical Integration Network for EEG and MEG. A residual sensor adapter unifies input dimensions, intermediate dilated-block states retain temporal depth, and a target- and time-dependent gate fuses local hierarchical and attention-enhanced context predictions. Across two EEG and two MEG datasets, SHINE has the highest mean envelope and mean-Mel Pearson correlations among nine local baseline implementations on all eight dataset-metric combinations. SHINE also placed second in the speech-detection Extended Track of the NeurIPS 2025 PNPL Competition. Code will be released at this https URL.
Sound (cs.SD); Artificial Intelligence (cs.AI)
arXiv:2602.23960 [cs.SD]
(or arXiv:2602.23960v2 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2602.23960 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 247 cuvinte
Subiecte conexe
Netskope (NTSK) Q2 2027 Earnings Call Transcript
fool.com · 10 sept. 2026
Graph-Based Inference and Topology-Aware Multi-Agent Reinforcement Learning for Large-Scale Railway Network Management
arXiv · 26 sept. 2026
IPGeoAI: Transformer-Based Geolocation with LLM Semantic Fusion
arXiv · 7 sept. 2026
Graph Mamba Survival Analysis Based on Topology-Aware ordering
arXiv · 3 iun. 2026