{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,5]],"date-time":"2025-11-05T07:02:09Z","timestamp":1762326129994,"version":"3.41.2"},"reference-count":45,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T00:00:00Z","timestamp":1750982400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neuroinform."],"abstract":"<jats:sec><jats:title>Introduction<\/jats:title><jats:p><jats:italic>Imagined speech<\/jats:italic> decoding using EEG holds promising applications for individuals with motor neuron diseases, although its performance remains limited due to small dataset sizes and the absence of sensory feedback. Here, we investigated whether incorporating EEG data from <jats:italic>overt<\/jats:italic> (pronounced) speech could enhance <jats:italic>imagined speech<\/jats:italic> classification.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Our approach systematically compares four classification scenarios by modifying the training dataset: intra-subject (using only <jats:italic>imagined speech<\/jats:italic>, combining <jats:italic>overt<\/jats:italic> and <jats:italic>imagined speech<\/jats:italic>, and using only <jats:italic>overt speech<\/jats:italic>) and multi-subject (combining <jats:italic>overt speech<\/jats:italic> data from different participants with the <jats:italic>imagined speech<\/jats:italic> of the target participant). We implemented all scenarios using the convolutional neural network EEGNet. To this end, twenty-four healthy participants pronounced and imagined five Spanish words.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>In binary word-pair classifications, combining <jats:italic>overt<\/jats:italic> and <jats:italic>imagined speech<\/jats:italic> data in the intra-subject scenario led to accuracy improvements of 3%\u20135.17% in four out of 10 word pairs, compared to training with <jats:italic>imagined speech<\/jats:italic> only. Although the highest individual accuracy (95%) was achieved with <jats:italic>imagined speech<\/jats:italic> alone, the inclusion of <jats:italic>overt speech<\/jats:italic> data allowed more participants to surpass 70% accuracy, increasing from 10 (<jats:italic>imagined only<\/jats:italic>) to 15 participants. In the intra-subject multi-class scenario, combining <jats:italic>overt<\/jats:italic> and <jats:italic>imagined speech<\/jats:italic> did not yield statistically significant improvements over using <jats:italic>imagined speech<\/jats:italic> exclusively.<\/jats:p><\/jats:sec><jats:sec><jats:title>Discussion<\/jats:title><jats:p>Finally, we observed that features such as word length, phonological complexity, and frequency of use contributed to higher discriminability between certain <jats:italic>imagined<\/jats:italic> word pairs. These findings suggest that incorporating <jats:italic>overt speech<\/jats:italic> data can improve <jats:italic>imagined speech<\/jats:italic> decoding in individualized models, offering a feasible strategy to support the early adoption of brain-computer interfaces before speech deterioration occurs in individuals with motor neuron diseases.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fninf.2025.1583428","type":"journal-article","created":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T05:31:36Z","timestamp":1751002296000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["From pronounced to imagined: improving speech decoding with multi-condition EEG data"],"prefix":"10.3389","volume":"19","author":[{"given":"Denise","family":"Alonso-V\u00e1zquez","sequence":"first","affiliation":[]},{"given":"Omar","family":"Mendoza-Montoya","sequence":"additional","affiliation":[]},{"given":"Ricardo","family":"Caraza","sequence":"additional","affiliation":[]},{"given":"Hector R.","family":"Martinez","sequence":"additional","affiliation":[]},{"given":"Javier M.","family":"Antelis","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,6,27]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"3001","DOI":"10.1007\/s11831-021-09684-6","article-title":"Review of machine learning techniques for EEG based brain computer interface","volume":"29","author":"Aggarwal","year":"2022","journal-title":"Arch. Comput. Methods Eng"},{"key":"B2","doi-asserted-by":"crossref","DOI":"10.52591\/lxai202307233","article-title":"\u201cRecognition of grammatical classes of imagined speech words using a convolutional neural network and brain signals,\u201d","volume-title":"LatinX in AI Workshop at ICML 2023 (Regular Deadline)","author":"Alonso-V\u00e1zquez","year":""},{"key":"B3","doi-asserted-by":"publisher","first-page":"225","DOI":"10.3390\/computation11110225","article-title":"EEG-based classification of spoken words using machine learning approaches","volume":"11","author":"Alonso-V\u00e1zquez","year":"","journal-title":"Computation"},{"key":"B4","first-page":"2390","article-title":"\u201cFilter bank common spatial pattern (FBCSP) in brain-computer interface,\u201d","volume-title":"2008 IEEE international joint conference on neural networks (IEEE world congress on computational intelligence)","author":"Ang","year":"2008"},{"key":"B5","doi-asserted-by":"publisher","first-page":"9617","DOI":"10.1038\/s41598-024-60277-2","article-title":"Online speech synthesis using a chronically implanted brain-computer interface in an individual with ALS","volume":"14","author":"Angrick","year":"2024","journal-title":"Sci. Rep"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1398065","DOI":"10.3389\/fnhum.2024.1398065","article-title":"Decoding imagined speech with delay differential analysis","volume":"18","author":"Carvalho","year":"2024","journal-title":"Front. Hum. Neurosci"},{"key":"B7","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1080\/24709360.2017.1396742","article-title":"A tutorial on kernel density estimation and recent advances","volume":"1","author":"Chen","year":"2017","journal-title":"Biostat. Epidemiol"},{"key":"B8","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1016\/j.jneumeth.2015.01.010","article-title":"Exceeding chance level by chance: the caveat of theoretical chance levels in brain signal classification and statistical assessment of decoding accuracy","volume":"250","author":"Combrisson","year":"2015","journal-title":"J. Neurosci. Methods"},{"key":"B9","first-page":"2782","article-title":"\u201cNeural speech decoding for amyotrophic lateral sclerosis,\u201d","volume-title":"INTERSPEECH","author":"Dash","year":""},{"key":"B10","doi-asserted-by":"publisher","first-page":"290","DOI":"10.3389\/fnins.2020.00290","article-title":"Decoding imagined and spoken phrases from non-invasive neural (MEG) signals","volume":"14","author":"Dash","year":"","journal-title":"Front. Neurosci"},{"key":"B11","doi-asserted-by":"publisher","first-page":"1114811","DOI":"10.3389\/fpsyg.2024.1114811","article-title":"Automatic detection of ALS from single-trial MEG signals during speech tasks: a pilot study","volume":"15","author":"Dash","year":"2024","journal-title":"Front. Psychol"},{"key":"B12","doi-asserted-by":"publisher","first-page":"301","DOI":"10.1016\/j.neucom.2021.08.035","article-title":"Recognition of grammatical class of imagined words from EEG signals using convolutional neural network","volume":"465","author":"Datta","year":"2021","journal-title":"Neurocomputing"},{"key":"B13","doi-asserted-by":"publisher","first-page":"15713","DOI":"10.1038\/s41598-022-19651-1","article-title":"Rate of speech decline in individuals with amyotrophic lateral sclerosis","volume":"12","author":"Eshghi","year":"2022","journal-title":"Sci. Rep"},{"key":"B14","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511817779","author":"Fitch","year":"2010","journal-title":"The Evolution of Language"},{"key":"B15","doi-asserted-by":"publisher","first-page":"e0291908","DOI":"10.1371\/journal.pone.0291908","article-title":"Challenges in the real world use of classification accuracy metrics: from recall and precision to the matthews correlation coefficient","volume":"18","author":"Foody","year":"2023","journal-title":"PLoS ONE"},{"key":"B16","doi-asserted-by":"publisher","first-page":"1580","DOI":"10.1016\/S1388-2457(03)00093-2","article-title":"EMG contamination of EEG: spectral and topographical characteristics","volume":"114","author":"Goncharova","year":"2003","journal-title":"Clin. Neurophysiol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"1383","DOI":"10.1016\/j.neuroimage.2005.11.048","article-title":"The time course of visual word recognition as revealed by linear regression analysis of ERP data","volume":"30","author":"Hauk","year":"2006","journal-title":"Neuroimage"},{"key":"B18","doi-asserted-by":"publisher","first-page":"2529","DOI":"10.1007\/s11517-024-03083-2","article-title":"Imagined speech classification exploiting EEG power spectrum features","volume":"62","author":"Hossain","year":"2024","journal-title":"Med. Biol. Eng. Comput"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511921803","author":"Japkowicz","year":"2011","journal-title":"Evaluating Learning Algorithms: A Classification Perspective"},{"key":"B20","doi-asserted-by":"publisher","first-page":"056013","DOI":"10.1088\/1741-2552\/aace8c","article-title":"Eegnet: a compact convolutional neural network for EEG-based brain-computer interfaces","volume":"15","author":"Lawhern","year":"2018","journal-title":"J. Neural Eng"},{"key":"B21","doi-asserted-by":"crossref","first-page":"4441","DOI":"10.1109\/SMC53992.2023.10394004","article-title":"\u201cToinet: transfer learning from overt speech-to imagined speech-based EEG signals with convolutional autoencoder,\u201d","volume-title":"2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","author":"Lee","year":"2023"},{"key":"B22","first-page":"387","article-title":"\u201cEEG representations of spatial and temporal features in imagined speech and overt speech,\u201d","volume-title":"Pattern Recognition: 5th Asian Conference, ACPR 2019, Auckland, New Zealand, November 26-29, 2019, Revised Selected Papers, Part II 5","author":"Lee","year":"2019"},{"key":"B23","doi-asserted-by":"publisher","first-page":"2647","DOI":"10.1109\/TNSRE.2020.3040289","article-title":"Neural decoding of imagined speech and visual imagery as intuitive paradigms for BCI communication","volume":"28","author":"Lee","year":"2020","journal-title":"IEEE Trans. Neural Syst. Rehabil. Eng"},{"key":"B24","doi-asserted-by":"publisher","DOI":"10.1136\/jnnp.74.suppl_4.iv32","article-title":"The management of motor neurone disease","author":"Leigh","year":"2003","journal-title":"J. Neurol. Neurosurg. Psychiatry"},{"key":"B25","doi-asserted-by":"publisher","first-page":"R1","DOI":"10.1088\/1741-2560\/4\/2\/R01","article-title":"A review of classification algorithms for EEG-based brain-computer interfaces","volume":"4","author":"Lotte","year":"2007","journal-title":"J. Neural Eng."},{"volume-title":"The Oxford Handbook of Event-Related Potential Components","year":"2013","author":"Luck","key":"B26"},{"key":"B27","doi-asserted-by":"publisher","first-page":"14","DOI":"10.3389\/fneng.2014.00014","article-title":"Decoding spectrotemporal features of overt and covert speech from the human cortex","volume":"7","author":"Martin","year":"2014","journal-title":"Front. Neuroeng"},{"key":"B28","doi-asserted-by":"publisher","first-page":"182","DOI":"10.1016\/j.ijpsycho.2013.03.017","article-title":"Brain correlates of cognitive inhibition in fibromyalgia: emotional intrusion of symptom-related words","volume":"88","author":"Mercado","year":"2013","journal-title":"Int. J. Psychophysiol"},{"key":"B29","doi-asserted-by":"publisher","first-page":"147778","DOI":"10.1016\/j.brainres.2022.147778","article-title":"A comparison and classification of oscillatory characteristics in speech perception and covert speech","volume":"1781","author":"Moon","year":"2022","journal-title":"Brain Res"},{"key":"B30","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1056\/NEJMoa2027540","article-title":"Neuroprosthesis for decoding speech in a paralyzed person with anarthria","volume":"385","author":"Moses","year":"2021","journal-title":"N. Engl. J. Med"},{"key":"B31","doi-asserted-by":"publisher","first-page":"016002","DOI":"10.1088\/1741-2552\/aa8235","article-title":"Inferring imagined speech using EEG signals: a new approach using riemannian manifold features","volume":"15","author":"Nguyen","year":"2017","journal-title":"J. Neural Eng"},{"key":"B32","doi-asserted-by":"publisher","first-page":"52","DOI":"10.1038\/s41597-022-01147-2","article-title":"Thinking out loud, an open-access EEG-based bci dataset for inner speech recognition","volume":"9","author":"Nieto","year":"2022","journal-title":"Sci. Data"},{"key":"B33","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s41467-021-27725-3","article-title":"Imagined speech can be decoded from low-and cross-frequency intracranial EEG features","volume":"13","author":"Proix","year":"2022","journal-title":"Nat. Commun"},{"key":"B34","doi-asserted-by":"publisher","first-page":"2910","DOI":"10.1007\/s00415-022-10971-w","article-title":"Emerging technologies for management of patients with amyotrophic lateral sclerosis: from telehealth to assistive robotics and neural interfaces","volume":"269","author":"Pugliese","year":"2022","journal-title":"J. Neurol"},{"key":"B35","doi-asserted-by":"crossref","first-page":"2650","DOI":"10.1109\/SMC53654.2022.9945447","article-title":"\u201cImproving silent speech bci training procedures through transfer from overt to silent speech,\u201d","volume-title":"2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","author":"Rekrut","year":"2022"},{"key":"B36","doi-asserted-by":"publisher","first-page":"2289","DOI":"10.1109\/TBME.2013.2253608","article-title":"Transferring subspaces between subjects in brain-computer interfacing","volume":"60","author":"Samek","year":"2013","journal-title":"IEEE Trans. Biomed. Eng"},{"key":"B37","doi-asserted-by":"publisher","first-page":"6503","DOI":"10.3390\/s21196503","article-title":"Recognition of EEG signals from imagined vowels using deep learning methods","volume":"21","author":"Sarmiento","year":"2021","journal-title":"Sensors"},{"key":"B38","doi-asserted-by":"publisher","first-page":"473","DOI":"10.1111\/1469-8986.3740473","article-title":"Electrophysiological estimates of the time course of semantic and phonological encoding during implicit picture naming","volume":"37","author":"Schmitt","year":"2000","journal-title":"Psychophysiology"},{"key":"B39","doi-asserted-by":"publisher","first-page":"397","DOI":"10.1093\/oxfordhb\/9780195374148.013.0197","article-title":"\u201cLanguage-related ERP components,\u201d","author":"Swaab","year":"2012","journal-title":"The Oxford Handbook of Event-related Potential Components"},{"key":"B40","doi-asserted-by":"publisher","first-page":"520","DOI":"10.1038\/381520a0","article-title":"Speed of processing in the human visual system","volume":"381","author":"Thorpe","year":"1996","journal-title":"Nature"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1185","DOI":"10.1212\/01.CON.0000455886.14298.a4","article-title":"Als and other motor neuron diseases","volume":"20","author":"Tiryaki","year":"2014","journal-title":"CONTINUUM"},{"key":"B42","doi-asserted-by":"publisher","first-page":"176","DOI":"10.1016\/j.ijpsycho.2011.09.015","article-title":"Prediction during language comprehension: benefits, costs, and erp components","volume":"83","author":"Van Petten","year":"2012","journal-title":"Int. J. Psychophysiol"},{"key":"B43","doi-asserted-by":"publisher","first-page":"6744","DOI":"10.3390\/s21206744","article-title":"Silent EEG-speech recognition using convolutional and recurrent neural network with 85% accuracy of 9 words classification","volume":"21","author":"Vorontsova","year":"2021","journal-title":"Sensors"},{"key":"B44","doi-asserted-by":"publisher","first-page":"80","DOI":"10.2307\/3001968","article-title":"Individual comparisons by ranking methods","volume":"1","author":"Wilcoxon","year":"1945","journal-title":"Biometrics"},{"key":"B45","doi-asserted-by":"publisher","first-page":"110713","DOI":"10.1016\/j.brainresbull.2023.110713","article-title":"Artificial intelligence based multimodal language decoding from brain activity: a review","volume":"201","author":"Zhao","year":"2023","journal-title":"Brain Res. Bull"}],"container-title":["Frontiers in Neuroinformatics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2025.1583428\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T05:31:38Z","timestamp":1751002298000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fninf.2025.1583428\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,27]]},"references-count":45,"alternative-id":["10.3389\/fninf.2025.1583428"],"URL":"https:\/\/doi.org\/10.3389\/fninf.2025.1583428","relation":{},"ISSN":["1662-5196"],"issn-type":[{"type":"electronic","value":"1662-5196"}],"subject":[],"published":{"date-parts":[[2025,6,27]]},"article-number":"1583428"}}