{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T14:17:11Z","timestamp":1788445031371,"version":"build-2803163510"},"reference-count":47,"publisher":"IOP Publishing","issue":"5","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/publishingsupport.iopscience.iop.org\/iop-standard\/v1"},{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["J. Neural Eng."],"published-print":{"date-parts":[[2026,10,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    <jats:italic>Objective.<\/jats:italic>\n                    Electroencephalography (EEG)-guided target speaker extraction aims to recover the listener\u2019s attended speech from mixed speech. However, effectively representing and integrating the target-related information shared between EEG and speech remains challenging. This study focuses on this challenge.\n                    <jats:italic>Approach.<\/jats:italic>\n                    We propose TriGA-Net, a Graph Attention Network for Brain-Controlled Speaker Extraction, in which EEG recorded from the listener is used to guide target speech extraction. The EEG encoder combines multi-scale temporal features and frequency-domain features with graph convolution to model dependencies among electrodes, while self-attention captures interactions across the full set of EEG channels. The resulting EEG representation is fused with encoded speech features and passed to a MossFormer2 separator. By combining MossFormer with a recurrent module that does not rely on recurrent neural networks, the separator models long-range context together with the rhythmic and prosodic structure of speech.\n                    <jats:italic>Main results.<\/jats:italic>\n                    Experiments on the public Cocktail Party and KU Leuven (KUL) datasets yielded scale-invariant signal-to-distortion ratio (SI-SDR) values of 15.91 and 16.90 dB, respectively. Compared with the strongest baseline on each dataset, TriGA-Net improved SI-SDR by 1.96 dB on the Cocktail Party dataset and by 2.30 dB on the KUL dataset. Additional improvements were observed in short-time objective intelligibility (STOI) and extended STOI (ESTOI) on both datasets and in perceptual evaluation of speech quality on the KUL dataset.\n                    <jats:italic>Significance.<\/jats:italic>\n                    These results suggest that jointly modeling temporal-frequency EEG information, inter-electrode relationships, and long-range speech structure is effective for EEG-guided target speaker extraction.\n                  <\/jats:p>","DOI":"10.1088\/1741-2552\/ae9e8f","type":"journal-article","created":{"date-parts":[[2026,8,25]],"date-time":"2026-08-25T22:54:41Z","timestamp":1787698481000},"page":"056005","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["TriGA-Net: a graph attention network for brain-controlled speaker extraction"],"prefix":"10.1088","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-9522-7893","authenticated-orcid":true,"given":"Youhao","family":"Si","sequence":"first","affiliation":[{"name":"College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuan","family":"Liao","sequence":"additional","affiliation":[{"name":"School of Data Science, The Chinese University of Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-2477-5379","authenticated-orcid":true,"given":"Qiushi","family":"Han","sequence":"additional","affiliation":[{"name":"College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rui","family":"Dai","sequence":"additional","affiliation":[{"name":"College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9194-3317","authenticated-orcid":true,"given":"Yuhang","family":"Yang","sequence":"additional","affiliation":[{"name":"College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1741-3752","authenticated-orcid":true,"given":"Liya","family":"Huang","sequence":"additional","affiliation":[{"name":"College of Electronic and Optical Engineering & College of Flexible Electronics (Future Technology), Nanjing University of Posts and Telecommunications"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"jneae9e8fbib1","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/ad867c","type":"journal-article","article-title":"Real-time control of a hearing instrument with EEG-based attention decoding","volume":"22","author":"Hjortkj\u00e6r","year":"2025","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib2","doi-asserted-by":"publisher","first-page":"668","DOI":"10.1109\/THMS.2022.3176212","type":"journal-article","article-title":"A neural-inspired architecture for EEG-based auditory attention detection","volume":"52","author":"Cai","year":"2022","journal-title":"IEEE Trans. Hum. Mach. Syst."},{"key":"jneae9e8fbib3","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1109\/ICASSP.2016.7471631","type":"conference-proceedings","article-title":"Deep clustering: discriminative embeddings for segmentation and separation","author":"Hershey","year":"2016"},{"key":"jneae9e8fbib4","doi-asserted-by":"publisher","first-page":"1256","DOI":"10.1109\/TASLP.2019.2915167","type":"journal-article","article-title":"Conv-TasNet: surpassing ideal time-frequency magnitude masking for speech separation","volume":"27","author":"Luo","year":"2019","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib5","doi-asserted-by":"publisher","first-page":"46","DOI":"10.1109\/ICASSP40776.2020.9054266","type":"conference-proceedings","article-title":"Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation","author":"Luo","year":"2020"},{"key":"jneae9e8fbib6","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/ICASSP39728.2021.9413901","type":"conference-proceedings","article-title":"Attention is all you need in speech separation","author":"Subakan","year":"2021"},{"key":"jneae9e8fbib7","doi-asserted-by":"publisher","first-page":"3221","DOI":"10.1109\/TASLP.2023.3304482","type":"journal-article","article-title":"TF-GridNet: integrating full- and sub-band modeling for speech separation","volume":"31","author":"Wang","year":"2023","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib8","doi-asserted-by":"publisher","first-page":"0460d6","DOI":"10.1088\/1741-2552\/ac16b4","type":"journal-article","article-title":"Ear-EEG-based binaural speech enhancement (ee-BSE) using auditory attention detection and audiometric characteristics of hearing-impaired subjects","volume":"18","author":"Geravanchizadeh","year":"2021","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib9","doi-asserted-by":"publisher","first-page":"1467","DOI":"10.1109\/LSP.2022.3175130","type":"journal-article","article-title":"Speaker extraction with co-speech gestures cue","volume":"29","author":"Pan","year":"2022","journal-title":"IEEE Signal Process. Lett."},{"key":"jneae9e8fbib10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICASSP49357.2023.10096664","type":"conference-proceedings","article-title":"Target speaker extraction with ultra-short reference speech by VE-VE framework","author":"Yang","year":"2023"},{"key":"jneae9e8fbib11","doi-asserted-by":"publisher","first-page":"3032","DOI":"10.1109\/TASLP.2022.3205759","type":"journal-article","article-title":"USEV: universal speaker extraction with visual cue","volume":"30","author":"Pan","year":"2022","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib12","doi-asserted-by":"publisher","first-page":"7287","DOI":"10.1109\/ICASSP43922.2022.9746221","type":"conference-proceedings","article-title":"L-SpEx: localized target speaker extraction","author":"Ge","year":"2022"},{"key":"jneae9e8fbib13","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1109\/MSP.2021.3075932","type":"journal-article","article-title":"Electroencephalography-based auditory attention decoding: toward neurosteered hearing devices","volume":"38","author":"Geirnaert","year":"2021","journal-title":"IEEE Signal Process. Mag."},{"key":"jneae9e8fbib14","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-37959-4","type":"journal-article","article-title":"Target of selective auditory attention can be robustly followed with MEG","volume":"13","author":"Kurmanavi\u010di\u016bt\u0117","year":"2023","journal-title":"Sci. Rep."},{"key":"jneae9e8fbib15","doi-asserted-by":"publisher","first-page":"3115","DOI":"10.24963\/ijcai.2024\/345","type":"conference-proceedings","article-title":"DBPNet: dual-branch parallel network with temporal-frequency fusion for auditory attention detection","author":"Ni","year":"2024"},{"key":"jneae9e8fbib16","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/aab2f2","type":"journal-article","article-title":"A review of classification algorithms for EEG-based brain-computer interfaces: a 10-year update","volume":"15","author":"Lotte","year":"2018","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib17","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2020.117282","type":"journal-article","article-title":"Brain-informed speech separation (BISS) for enhancement of target speaker in multitalker speech perception","volume":"223","author":"Ceolini","year":"2020","journal-title":"NeuroImage"},{"key":"jneae9e8fbib18","doi-asserted-by":"publisher","first-page":"1310","DOI":"10.1109\/ICASSP39728.2021.9414969","type":"conference-proceedings","article-title":"Speaker-independent brain enhanced speech denoising","author":"Hosseini","year":"2021"},{"key":"jneae9e8fbib19","doi-asserted-by":"publisher","first-page":"1718","DOI":"10.1109\/TASLP.2022.3169629","type":"journal-article","article-title":"End-to-end brain-driven speech enhancement in multi-talker conditions","volume":"30","author":"Hosseini","year":"2022","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib20","doi-asserted-by":"publisher","first-page":"3117","DOI":"10.21437\/Interspeech.2023-673","type":"conference-proceedings","article-title":"BASEN: time-domain brain-assisted speech enhancement network with convolutional cross attention in multi-talker conditions","author":"Zhang","year":"2023"},{"key":"jneae9e8fbib21","doi-asserted-by":"publisher","first-page":"4456","DOI":"10.1109\/TASLP.2024.3463498","type":"journal-article","article-title":"NeuroHeed: neuro-steered speaker extraction using EEG signals","volume":"32","author":"Pan","year":"2024","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib22","doi-asserted-by":"publisher","first-page":"11456","DOI":"10.1109\/ICASSP48485.2024.10446333","type":"conference-proceedings","article-title":"NeuroHeed+: improving neuro-steered speaker extraction with joint auditory attention detection","author":"Pan","year":"2024"},{"key":"jneae9e8fbib23","doi-asserted-by":"publisher","first-page":"8040","DOI":"10.24963\/ijcai.2025\/894","type":"conference-proceedings","article-title":"M3ANet: multi-scale and multi-modal alignment network for brain-assisted target speaker extraction","author":"Fan","year":"2025"},{"key":"jneae9e8fbib24","doi-asserted-by":"publisher","first-page":"212","DOI":"10.1016\/j.neuroscience.2025.11.040","type":"journal-article","article-title":"GCANet: enhancing EEG-based auditory attention decoding with temporal frequency GCN and cross-attention mechanisms","volume":"593","author":"Dai","year":"2026","journal-title":"Neuroscience"},{"key":"jneae9e8fbib25","article-title":"TFGA-Net: temporal-frequency graph attention network for brain-controlled speaker extraction","author":"Si","year":"2025","type":"preprint"},{"key":"jneae9e8fbib26","doi-asserted-by":"publisher","first-page":"12189","DOI":"10.1109\/TCYB.2021.3071860","type":"journal-article","article-title":"Spatio-temporal-spectral hierarchical graph convolutional network with semisupervised active learning for patient-specific seizure prediction","volume":"52","author":"Li","year":"2022","journal-title":"IEEE Trans. Cybern."},{"key":"jneae9e8fbib27","doi-asserted-by":"publisher","first-page":"2218","DOI":"10.1109\/TAFFC.2022.3199075","type":"journal-article","article-title":"A dual-branch dynamic graph convolution based adaptive transformer feature fusion network for EEG emotion recognition","volume":"13","author":"Sun","year":"2022","journal-title":"IEEE Trans. Affect. Comput."},{"key":"jneae9e8fbib28","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/ac975c","type":"journal-article","article-title":"Detecting the locus of auditory attention based on the spectro-spatial-temporal analysis of EEG","volume":"19","author":"Jiang","year":"2022","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib29","doi-asserted-by":"publisher","DOI":"10.1016\/j.compbiomed.2023.107126","type":"journal-article","article-title":"A multi-head residual connection GCN for EEG emotion recognition","volume":"163","author":"Qiu","year":"2023","journal-title":"Comput. Biol. Med."},{"key":"jneae9e8fbib30","doi-asserted-by":"publisher","DOI":"10.3389\/fninf.2022.844667","type":"journal-article","article-title":"On the minimal amount of EEG data required for learning distinctive human features for task-dependent biometric applications","volume":"16","author":"G\u00f3mez-tapia","year":"2022","journal-title":"Front. Neuroinform."},{"key":"jneae9e8fbib31","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","type":"journal-article","article-title":"Graph neural networks: a review of methods and applications","volume":"1","author":"Zhou","year":"2020","journal-title":"AI Open"},{"key":"jneae9e8fbib32","first-page":"5998","type":"journal-article","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"jneae9e8fbib33","article-title":"An empirical evaluation of generic convolutional and recurrent networks for sequence modeling","author":"Bai","year":"2018","type":"preprint"},{"key":"jneae9e8fbib34","doi-asserted-by":"publisher","first-page":"10356","DOI":"10.1109\/ICASSP48485.2024.10445985","type":"conference-proceedings","article-title":"MossFormer2: combining transformer and RNN-free recurrent network for enhanced time-domain monaural speech separation","author":"Zhao","year":"2024"},{"key":"jneae9e8fbib35","doi-asserted-by":"publisher","first-page":"803","DOI":"10.1016\/j.cub.2018.01.080","type":"journal-article","article-title":"Electrophysiological correlates of semantic dissimilarity reflect the comprehension of natural, narrative speech","volume":"28","author":"Broderick","year":"2018","journal-title":"Curr. Biol."},{"key":"jneae9e8fbib36","doi-asserted-by":"publisher","first-page":"1697","DOI":"10.1093\/cercor\/bht355","type":"journal-article","article-title":"Attentional selection in a cocktail party environment can be decoded from single-trial EEG","volume":"25","author":"O\u2019Sullivan","year":"2015","journal-title":"Cereb. Cortex"},{"key":"jneae9e8fbib37","doi-asserted-by":"publisher","first-page":"402","DOI":"10.1109\/TNSRE.2016.2571900","type":"journal-article","article-title":"Auditory-inspired speech envelope extraction methods for improved EEG-based auditory attention detection in a cocktail party scenario","volume":"25","author":"Biesmans","year":"2017","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng."},{"key":"jneae9e8fbib38","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2560\/13\/5\/056014","type":"journal-article","article-title":"The effect of head-related filtering and ear-specific decoding bias on auditory attention detection","volume":"13","author":"Das","year":"2016","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib39","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1016\/j.jneumeth.2003.10.009","type":"journal-article","article-title":"EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis","volume":"134","author":"Delorme","year":"2004","journal-title":"J. Neurosci. Methods."},{"key":"jneae9e8fbib40","first-page":"4214","type":"conference-proceedings","article-title":"A short-time objective intelligibility measure for time-frequency weighted noisy speech","author":"Taal","year":"2010"},{"key":"jneae9e8fbib41","doi-asserted-by":"publisher","first-page":"2009","DOI":"10.1109\/TASLP.2016.2585878","type":"journal-article","article-title":"An algorithm for predicting the intelligibility of speech masked by modulated noise maskers","volume":"24","author":"Jensen","year":"2016","journal-title":"IEEE\/ACM Trans. Audio Speech Lang. Process."},{"key":"jneae9e8fbib42","first-page":"749","type":"conference-proceedings","article-title":"Perceptual evaluation of speech quality (PESQ)\u2014a new method for speech quality assessment of telephone networks and codecs","volume":"vol 2","author":"Rix","year":"2001"},{"key":"jneae9e8fbib43","first-page":"4190","type":"conference-proceedings","article-title":"Quantifying attention flow in transformers","author":"Abnar","year":"2020"},{"key":"jneae9e8fbib44","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/ace73f","type":"journal-article","article-title":"Relating EEG to continuous speech using deep neural networks: a review","volume":"20","author":"Puffay","year":"2023","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib45","doi-asserted-by":"publisher","DOI":"10.1088\/1741-2552\/ad2214","type":"journal-article","article-title":"What are we really decoding? Unveiling biases in EEG-based decoding of the spatial focus of auditory attention","volume":"21","author":"Rotaru","year":"2024","journal-title":"J. Neural Eng."},{"key":"jneae9e8fbib46","doi-asserted-by":"publisher","unstructured":"https:\/\/doi.org\/10.5061\/dryad.070jc ;","DOI":"10.5061\/dryad.070jc;","type":"dataset"},{"key":"jneae9e8fbib46","doi-asserted-by":"publisher","unstructured":"https:\/\/doi.org\/10.5281\/zenodo.4004271","DOI":"10.5281\/zenodo.4004271","type":"dataset"}],"container-title":["Journal of Neural Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T13:49:20Z","timestamp":1788443360000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/1741-2552\/ae9e8f"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":47,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2026,9,3]]},"published-print":{"date-parts":[[2026,10,1]]}},"URL":"https:\/\/doi.org\/10.1088\/1741-2552\/ae9e8f","relation":{},"ISSN":["1741-2560","1741-2552"],"issn-type":[{"value":"1741-2560","type":"print"},{"value":"1741-2552","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]},"assertion":[{"value":"TriGA-Net: a graph attention network for brain-controlled speaker extraction","name":"article_title","label":"Article Title"},{"value":"Journal of Neural Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2026 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.","name":"copyright_information","label":"Copyright Information"},{"value":"2026-04-13","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-08-25","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2026-09-03","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}