[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117623-en":3,"doc-seo-117623-105":30,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117623,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Decoding Melodic Acoustic Features from Neural Data","Decoding melodic acoustic features from EEG signals using machine learning to map neural responses to music envelope and onset characteristics, with evaluation across EEG frequency bands. The study compares musicians and non-musicians, showing higher decoding accuracy for musicians, with the theta band (4–8 Hz) yielding the best overall performance. Envelope decoding outperforms onset decoding, and results suggest measurable differences in neural representation of music that can inform neural audio engineering, assistive technologies, and neurological rehabilitation.","Audio Engineering Society  \nLate Breaking Demo Paper  \nPresented at the AES International Conference on  \nArtificial Intelligence and Machine Learning for Audio  \n2025 September 8–10, London, UK  \nThis Late Breaking Demo Paper was selected after a minimal screening process and was not peer reviewed. This Paper has been reproduced from the author's advance manuscript without editing, corrections, or consideration by the Review Board.  \nThe AES takes no responsibility for the contents. Reproduction of this paper, or any portion thereof, is not permitted without direct permission from the Audio Engineering Society.  \nDecoding Melodic Acoustic Features from Neural Data  \nZorka Bozilovic 1 and Iran R. Roman 1  \n1 Queen Mary University of London  \nCorrespondence should be addressed to Zorka Bozilovic ( [zorka.bozilovic@gmail.com](zorka.bozilovic@gmail.com))  \nABSTRACT  \nWe decoded acoustic features of music from EEG using machine learning, focusing on envelopes and onsetsand comparing musicians and non-musicians. Results showed higher accuracy when decoding from musicians, especially in the theta band. These findings highlight measurable differences in neural representation of music and suggest potential applications in neural audio engineering, assistive technologies, and rehabilitation.  \n1 Introduction  \nThis late-breaking paper investigates how auditory stimuli relate to brain activity by decoding acoustic features of music from EEG. This deepens understanding of how music transforms into neural signals and suggests directions for neural audio engineering (e.g., reconstructing imagined sounds) which may benefit assistive technologies and neurological rehabilitation [1] . Brain encoding and decoding have both been studied with audio [2]: encoding predicts EEG, while decoding reconstructs stimuli. A recent study examined musical feature encoding across EEG frequency bands [3], but did not explore decoding. We applied machine learning to a public EEG dataset recorded as participants listened to melodies. Our paper's contributions are: (1) Decoded melodic envelope and onsets from EEG; (2) Evaluated decoding across musicians vs. nonmusicians and EEG bands; (3) Open-sourced the code.1  \n1 github/ZorkaBozilovic/audio-brain-decoding  \n2 Methods  \nDataset. We used the dataset from [4](analyzed in [3]), which includes EEG (64ch) from 10 non-musicians and 10 pianists. Each individual passively listened to 10 melodies while EEG was recorded. This was repeated three times, yielding 30 trials per participant. EEG was re-referenced to the mean of the two mastoid channels.  \nPreprocessing. EEG was filtered into delta (1–4 Hz), theta (4–8 Hz), alpha (8–12 Hz), and beta (12–30 Hz) bands2. We extracted the envelopes and onset features from the dataset's melodies.  \nDecoding acoustic features from neural data. We used the temporal response function (TRF, a linear model mapping neural responses to stimulus features via convolution [2]) . When decoding, EEG is the input and musical features are the output. One model was fit  \n2Gamma was excluded due to the sampling rate in the pipeline [2] .  \nper participant, with separate models for envelope and onsets. We used eelbrain's boosting function with a 300ms window and evaluated decoding accuracy using Pearson's r between predicted and actual features.  \n3 Results  \nFigure 1 shows decoding performance across frequency bands. Correlation coefficients were significantly higher for musicians than non-musicians, with the theta band (4–8 Hz) yielding best decoding overall. Envelope decoding generally outperformed onsets decoding.  \nFig. 1: Decoding correlation coefficients across EEG bands for envelopes (left) & onsets (right) . The mean is shown in red. T-tests show significant differences across non-musicians & musicians (***: p \u003C 0.001, **: p \u003C 0.01, *: p \u003C 0.05) .  \nFigure 2 shows that musicians had stronger filter amplitudes and stronger voltage field, particularly over right-lateralized channels. This aligns with ","cbCaiesWiNQW9eCM","https://ap.wps.com/l/cbCaiesWiNQW9eCM","pdf",261840,1,2,"English","en",105,"# Abstract\n# Introduction\n# Methods\n## Dataset\n## Preprocessing\n## Decoding acoustic features from neural data\n# Results\n# Conclusion\n# References","[{\"question\":\"What musical acoustic features were decoded from EEG in this paper?\",\"answer\":\"The study decoded melodic acoustic features including envelopes and onsets from EEG signals using machine learning models.\"},{\"question\":\"How did performance differ between musicians and non-musicians?\",\"answer\":\"Decoding accuracy was higher for musicians than for non-musicians, with musicians showing significantly stronger correlation results.\"},{\"question\":\"Which EEG frequency band produced the best decoding results?\",\"answer\":\"The theta band (4–8 Hz) produced the best overall decoding performance, and envelope decoding generally outperformed onset decoding.\"}]","Decoding Melodic Acoustic Features from Neural Data | PDF",1785677343,5,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":28},"decoding-melodic-acoustic-features-from-neural-data","",{"@graph":36,"@context":84},[37,53,67],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":21},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/decoding-melodic-acoustic-features-from-neural-data/117623/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What musical acoustic features were decoded from EEG in this paper?","Question",{"text":74,"@type":75},"The study decoded melodic acoustic features including envelopes and onsets from EEG signals using machine learning models.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How did performance differ between musicians and non-musicians?",{"text":79,"@type":75},"Decoding accuracy was higher for musicians than for non-musicians, with musicians showing significantly stronger correlation results.",{"name":81,"@type":72,"acceptedAnswer":82},"Which EEG frequency band produced the best decoding results?",{"text":83,"@type":75},"The theta band (4–8 Hz) produced the best overall decoding performance, and envelope decoding generally outperformed onset decoding.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,108,113,118,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":29,"doc_module":4,"doc_module_name":46,"category_name":105,"show_sort_weight":106,"slug":107},"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":46,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":46,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":46,"category_name":135,"show_sort_weight":29,"slug":136},19,"General","general"]