On this page, you can find an overview of publicly available speech decoding datasets, that generally contain neural data of subjects listening to natural speech. This can be either with multiple competing talkers (to decode selective auditory attention) or with a single talker.
In case your publicly available dataset is not mentioned, please contact me.
Currently, there is about:
publicly available.
Multi-talker (selective attention)
| Original reference | Number of participants | Participant population | Amount of data per participant | Neurorecording system | Stimuli | Sex of competing talkers | Location of competing talkers | Acoustic room conditions | Comments | Link dataset | Link paper |
|---|---|---|---|---|---|---|---|---|---|---|---|
| W. Biesmans, N. Das, T. Francart, and A. Bertrand, “Auditory-Inspired Speech Envelope Extraction Methods for Improved EEG-Based Auditory Attention Detection in a Cocktail Party Scenario,” IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 25, no. 5, pp. 402–412, 2017 | 16 | young, normal hearing | 72 min (8 trials x 6 min + 12 trials x 2 min) | EEG 64-channel BioSemi | Dutch short stories | male-male | 90/-90 degrees | dichotic and HRTF-filtered in anechoic room | Dataset | Paper | |
| S. A. Fuglsang, T. Dau, and J. Hjortkjær, “Noise-robust cortical tracking of attended speech in real-world acoustic scenes,” NeuroImage, vol. 156, pp. 435–444, 2017 | 18 | young, normal hearing | 50 min (60 trials x 50 s) | EEG 64-channel BioSemi | Danish fictional stories | male-female | 60/-60 degrees | HRTF-filtered in anechoic, mildly, and highly reverberant room | EOG available | Dataset | Paper |
| S. A. Fuglsang, J. Märcher-Rørsted, T. Dau, J. Hjortkjær, "Effects of Sensorineural Hearing Loss on Cortical Synchronization to Competing Speech during Selective Attention," Journal of Neuroscience, vol. 40, no. 12, pp. 2562-2572, 2020 | 44 | 22 hearing impaired + 22 normal hearing | 26.7 min (32 trials x 50 s) | EEG 64-channel BioSemi | Danish audiobooks | male-female | 90/-90 degrees | HRTF-filtered | single-talker, ERPs, EFRs, resting-state also available / in-ear EEG for 19 of 44 participants / EOG available | Dataset | Paper |
| A. J. Power, J. J. Foxe, E.-J. Forde, R. B. Reilly, and E. C. Lalor, “At what time is the cocktail party? A late locus of selective attention to natural speech,” European Journal of Neuroscience, vol. 35, pp. 1497–1503, 2012 | 33 | young, normal hearing | 30 min (30 trials x 1 min) | EEG 128-channel BioSemi | English fictional stories | male-male | 90/-90 degrees | dichotic | used in seminal O'Sullivan paper | Dataset | Paper |
| A. Mundanad Narayanan, R. Zink, and A. Bertrand, “EEG miniaturization limits for stimulus decoding with EEG sensor networks,” Journal of Neural Engineering, vol. 18, no. 5, p. 056042, 2021 | 30 | young, normal hearing | 24 min (4 trials x 6 min) | EEG 255-channel SynAmps RT | Dutch fictional stories | male-male | 90/-90 degrees | HRTF-filtered | Dataset | Paper | |
| L. Straetmans, B. Holtze, S. Debener, M. Jaeger, and B. Mirkovic, “Neural tracking to go: auditory attention decoding and saliency detection with mobile EEG,” Journal of Neural Engineering, vol. 18, no. 6, p. 066054, 2021 | 20 | young, normal hearing | 30 min (6 trials x 5 min) | EEG 24-channel EasyCap GmbH/SMARTING | German audiobooks + natural salient events | male-male | 45/-45 degrees | HRTF-filtered, recorded in public cafeteria without other people | 3 trials during walking, 3 trials sitting / salient environmental sounds added | Dataset | Paper |
| S. Akram, A. Presacco, J. Z. Simon, S. A. Shamma, and B. Babadi, “Robust decoding of selective auditory attention from MEG in a competingspeaker environment via state-space modeling,” NeuroImage, vol. 124, pp. 906–917, 2016 | 7 | young, normal hearing | 6 min (2 conditions x 3 repetitions x 1 min) | MEG 157-channel | English fictional stories | male-female | 90/-90 degrees | dichotic | instructed attention switches 1 time per trial | Dataset | Paper |
| A. Presacco, S. Miran, B. Babadi, and J. Z. Simon, “Real-Time Tracking of Magnetoencephalographic Neuromarkers during a Dynamic Attention-Switching Task,” in Proceedings of the 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 4148–4151, 2019 | 5 | young, normal hearing | 4.5 min (3 trials x 90 s) | MEG 157-channel | English fictional stories | male-female | 90/-90 degrees | dichotic | at-will attention switches 1-3 times per trial | Dataset | Paper |
| C. Brodbeck, L. E. Hong, and J. Z. Simon, “Rapid Transformation from Auditory to Linguistic Representations of Continuous Speech,” Current Biology, vol. 28, no. 24, pp. 3976–3983.e5, 2018 | 26 | normal hearing | 16 min (4 trials x 4 repetitions x 1 min) | MEG 157-channel | English audiobooks | male-female | NA | NA | Dataset | Paper | |
| G. Cantisani, G. Trégoat, S. Essid, G. Richard, "MAD-EEG: an EEG dataset for decoding auditory attention to a target instrument in polyphonic music," Speech, Music and Mind (SMM), Satellite Workshop of Interspeech 2019, Vienna, Austria, 2019 | 8 | young, normal hearing, non-professional musicians | 30-32 min (78 stimuli x 4 repetitions x 6 s) | EEG 20-channel B-Alert X24 headset | Polyphonic music mixture (14 solos, 40 duets, 24 trios) | various instruments | NA | Speakers at 45/-45 degrees, convex weighting of instruments in the mixture | EOG, EMG, ECG, head motion acceleration available / single instrument available | Dataset | Paper |
| O. Etard, R. B. Messaoud, G. Gaugain, and T. Reichenbach, “No Evidence of Attentional Modulation of the Neural Response to the Temporal Fine Structure of Continuous Musical Pieces,” Journal of Cognitive Neuroscience, vol. 34, no. 3, pp. 411-424, 2022 | 17 | young, normal hearing | 22.4 min (7 stimuli of 11.2 min in total x 2 repetitions) | EEG 4-channel Ag/AgCl electrodes (Multitrode, BrainProducts) | Music (Bach's Two-Part Inventions) | piano-guitar | NA | dichotic | single instrument available | Dataset | Paper |
| Y. Zhang, H. Ruan, Z. Yuan, H. Du, X. Gao, and J. Lu, "A Learnable Spatial Mapping for Decoding the Directional Focus of Auditory Attention Using EEG," 2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Rhodes Island, Greece, pp. 1-5, 2023 | 21 | normal hearing | 64 min (32 trials x 2 min) | EEG 32-channel EMOTIV Epoc Flex Saline | Chinese news programs | male-female | random pairs from +-135/120/90/60/45/30/15 degrees | loudspeaker array | per trial, random pairs of the competing speaker directions are taken | Dataset | Paper |
| O. Etard, M. Kegler, C. Braiman, A.E. Forte, and T. Reichenbach, “Decoding of selective attention to continuous speech from the human auditory brainstem response,” NeuroImage, vol. 200, pp. 1-11, 2019 | 18 | young, normal hearing | 20 min (2 trials x 4 parts x 2.5 min) | EEG 64-channel actiCAP | English audiobooks | male-female | 90/-90 degrees | dichotic | Dataset | Paper | |
| I. Rotaru, S. Geirnaert, N. Heintz, I. Van de Ryck, A. Bertrand, and T. Francart, "What are we really decoding? Unveiling biases in EEG-based decoding of the spatial focus of auditory attention," Journal of Neural Engineering, vol. 21, no. 1, 016017, 2024 | 13 | young, normal hearing | 80 min (2 blocks x 4 conditions x 10 min) | EEG 64-channel BioSemi | Dutch science-outreach podcasts | male-male | 90/-90 degrees | HRTF-filtered in anechoic room | Per condition, a different audio-visual condition is used (moving video, moving target noise, no visuals, static video). EOG also available. Per trial (=condition), there is one switch in attention after 5 minutes | Dataset | Paper |
| Z. Lin, T. He, S. Cai, and H. Li, "ASA: An Auditory Spatial Attention Dataset with Multiple Speaking Locations," Interspeech 2024, Kos, Greece, 2024 | 20 | normal hearing | 24 min (20 trials x 1-1.5 min) | EEG 64-channel Easycap | Mandarin stories | male-female | 90/-90, -60/60, -45/45, -30/30, -5/5 degrees | HRTF-filtered through headphones | Dataset | Paper | |
| Y. Yan, X. Xu, H. Zhu, P. Tian, Z. Ge, X. Wu, and J. Chen, "Auditory Attention Decoding in Four-Talker Environment with EEG," Interspeech 2024, Kos, Greece, pp. 432-436, 2024 and H. Zhu et al., "Using Ear-EEG to Decode Auditory Attention in Multiple-speaker Environment," 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Hyderabad, India, pp. 1-5, 2025 | 16 | young, normal hearing | 40 min (40 trials x 1 min) | EEG 64-channel NeuSen Wireless + 20-channel cEEGrid | Mandarin stories | male (x2)-female (x2) | -90/-30/+30/+90 | loudspeaker array | 4-competing speaker paradigm; original stimuli available upon request (2301111611@stu.pku.edu.cn), only envelopes available | Dataset | Paper |
| M. Thornton, D. Mandic, and T. Reichenbach, "Decoding of selective attention to speech from Ear-EEG recordings," , arXiv, 2024 (arXiv:2401.05187) | 18 | young, normal hearing | 40 min (16 trials x 150 s) | EEG 2-channel in-ear (reference FT7) | English audiobooks | male-female | NA | diotic via headphones | Dataset | Paper | |
| Q. Wang, Q. Zhou, Z. Ma, N. Wang, T. Zhang, Y. Fu, and J. Li, "Le Petit Prince (LPP) multi-talker: Naturalistic 7 T fMRI and EEG dataset," Scientific Data,vol. 12, no. 829, 2025 | 25 | young | 20 min (2 trials x 10 min) | EEG 64-channel actiCAP | Mandarin audiobook ("The Little Prince") | male-female (synthesized) | NA | insert earphones | fixation on crosshair; also fMRI available and single-talker data | Dataset | Paper |
Single talker
| Original reference | Number of participants | Participant population | Amount of data per participant | Neurorecording system | Stimuli | Sex of talker | Comments | Link dataset | Link paper |
|---|---|---|---|---|---|---|---|---|---|
| G. M. Di Liberto, J. A. O’Sullivan, and E. C. Lalor, “Low-Frequency Cortical Entrainment to Speech Reflects Phoneme-Level Processing,” Current Biology, vol. 25, no. 19, pp. 2457–2465, 2015 and M. P. Broderick, A. J. Anderson, G. M. Di Liberto, M. J. Crosse, and E. C. Lalor, “Electrophysiological Correlates of Semantic Dissimilarity Reflect the Comprehension of Natural, Narrative Speech,” Current Biology, vol. 28, no. 5, pp. 803–809.e3, 2018 | 19 | young, normal-hearing | 60 min (20 trials x 180 s) | EEG 128-channel BioSemi | English fictional stories | male | Dataset | Paper | |
| G. M. Di Liberto, J. A. O’Sullivan, and E. C. Lalor, “Low-Frequency Cortical Entrainment to Speech Reflects Phoneme-Level Processing,” Current Biology, vol. 25, no. 19, pp. 2457–2465, 2015 | 10 | young, normal-hearing | 72.3 min (28 trials x 155 s) | EEG 128-channel BioSemi | English fictional stories, reversed | male | same stimuli as dataset above, but reversed | Dataset | Paper |
| H. Weissbart, K. D. Kandylaki, and T. Reichenbach, “Cortical Tracking of Surprisal during Continuous Speech Comprehension,” Journal of Cognitive Neuroscience, vol. 32, no. 1, pp. 155–166, 2020 | 13 | young, normal-hearing | 40 min (15 trials x approx. 2.6 min) | EEG 64-channel actiCAP | English short stories | male | Dataset | Paper | |
| F. J. Vanheusden, M. Kegler, K. Ireland, C. Georga, D. M. Simpson, T. Reichenbach, and S. L. Bell, “Hearing Aids Do Not Alter Cortical Entrainment to Speech at Audible Levels in Mild-to-Moderately Hearing- Impaired Subjects,” Frontiers in Human Neuroscience, vol. 14, no. 109, 2020 | 17 | older, hearing impaired, hearing aid users | 25 min (8 trials x approx. 3 min) | EEG 32-channel BioSemi | English audiobook | female | trials aided and unaided by hearing aid | Dataset | Paper |
| L. Bollens, B. Accou, H. Van hamme, and T. Francart, "A Large Auditory EEG Decoding Dataset", KU Leuven RDR, 2023 | 85 | young, normal-hearing | 130-150 min (8 -10 trials x 15 min) | EEG 64-channel BioSemi | Flemish audiobooks and podcasts | male and female | Dataset | ||
| J. R. Brennan, and J. T. Hale, "Hierarchical structure guides rapid linguistic predictions during naturalistic listening," PLoS ONE, vol. 14, no. 1, e0207741, 2019 | 49 | young | 12.4 min | EEG 61-channel actiCAP | English audiobook | female | no mention of medical conditions of participants | Dataset | Paper |
| S. A. Fuglsang, J. Märcher-Rørsted, T. Dau, J. Hjortkjær, "Effects of Sensorineural Hearing Loss on Cortical Synchronization to Competing Speech during Selective Attention," Journal of Neuroscience, vol. 40, no. 12, pp. 2562-2572, 2020 | 44 | 22 hearing impaired + 22 normal hearing | 13.3 min (16 trials x 50 s) | EEG 64-channel BioSemi | Danish audiobooks | male and female | dual-talker, ERPs, EFRs, resting-state also available / in-ear EEG for 19 of 44 participants | Dataset | Paper |
| L. Gwilliams, J.R. King, Marantz, A. Marantz, and D. Poeppel, "Neural dynamics of phoneme sequences reveal position-invariant code for content and order," Nature Communications, 13, 6606, 2022 | 27 | young, normal-hearing | 120 min (2 sessions x 1 hour) | MEG 208-channel | English fictional stories | Dataset | Paper | ||
| N. H. L. Lam, A. Hultén, P. Hagoort, and J.-M. Schoffelen, "Robust neuronal oscillatory entrainment to speech displays individual variation in lateralisation," Language, Cognition and Neuroscience, no. 33, vol. 8, pp. 943-954, 2018 + various other papers | 102 | young, healthy | around 8.4 min (120 sentences x 2.8-6 s) | MEG 275-channel | Dutch sentences | female | fMRI also available. Also resting-state and reading available | Dataset | Paper |
| C. Brodbeck, L. E. Hong, and J. Z. Simon, “Rapid Transformation from Auditory to Linguistic Representations of Continuous Speech,” Current Biology, vol. 28, no. 24, pp. 3976–3983.e5, 2018 | 26 | normal-hearing | 8 min (8 trials x 1 min) | MEG 157-channel | English audiobooks | male and female | Dataset | Paper | |
| O. Etard, and T. Reichenbach, "Neural speech tracking in the theta and in the delta frequency band differentially encode clarity and comprehension of speech in noise," Journal of Neuroscience, vol. 39, no. 29, pp. 5750-5759, 2019 | 12 | young, normal-hearing | 40 min (4 noise levels x 4 parts x 2.5 min) | EEG 64-channel actiCAP | English audiobooks | male and female | 4 levels of babble noise. Also EEG data of participantss listening to Dutch available (0% speech comprehension) | Dataset | Paper |
| Q. Wang, Q. Zhou, Z. Ma, N. Wang, T. Zhang, Y. Fu, and J. Li, "Le Petit Prince (LPP) multi-talker: Naturalistic 7 T fMRI and EEG dataset," Scientific Data,vol. 12, no. 829, 2025 | 25 | young | 20 min (2 trials x 10 min) | EEG 64-channel actiCAP | Mandarin audiobook ("The Little Prince") | male and female (synthesized) | also dual-speaker data available | Dataset | Paper |