Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387378260> ?p ?o ?g. }
- W4387378260 abstract "Abstract Decoding speech from brain activity is a long-awaited goal in both healthcare and neuroscience. Invasive devices have recently led to major milestones in this regard: deep-learning algorithms trained on intracranial recordings can now start to decode elementary linguistic features such as letters, words and audio-spectrograms. However, extending this approach to natural speech and non-invasive brain recordings remains a major challenge. Here we introduce a model trained with contrastive learning to decode self-supervised representations of perceived speech from the non-invasive recordings of a large cohort of healthy individuals. To evaluate this approach, we curate and integrate four public datasets, encompassing 175 volunteers recorded with magneto-encephalography or electro-encephalography while they listened to short stories and isolated sentences. The results show that our model can identify, from 3 seconds of magneto-encephalography signals, the corresponding speech segment with up to 41% accuracy out of more than 1,000 distinct possibilities on average across participants, and with up to 80% in the best participants—a performance that allows the decoding of words and phrases absent from the training set. The comparison of our model with a variety of baselines highlights the importance of a contrastive objective, pretrained representations of speech and a common convolutional architecture simultaneously trained across multiple participants. Finally, the analysis of the decoder’s predictions suggests that they primarily depend on lexical and contextual semantic representations. Overall, this effective decoding of perceived speech from non-invasive recordings delineates a promising path to decode language from brain activity, without putting patients at risk of brain surgery." @default.
- W4387378260 created "2023-10-06" @default.
- W4387378260 creator A5024428666 @default.
- W4387378260 creator A5048065823 @default.
- W4387378260 creator A5056821038 @default.
- W4387378260 creator A5075329396 @default.
- W4387378260 creator A5087825391 @default.
- W4387378260 date "2023-10-05" @default.
- W4387378260 modified "2023-10-16" @default.
- W4387378260 title "Decoding speech perception from non-invasive brain recordings" @default.
- W4387378260 cites W1488659701 @default.
- W4387378260 cites W1606214964 @default.
- W4387378260 cites W1819242701 @default.
- W4387378260 cites W1965248225 @default.
- W4387378260 cites W1981180290 @default.
- W4387378260 cites W2005066437 @default.
- W4387378260 cites W2015460481 @default.
- W4387378260 cites W2031186925 @default.
- W4387378260 cites W2036084760 @default.
- W4387378260 cites W2048425229 @default.
- W4387378260 cites W2049363352 @default.
- W4387378260 cites W2084333685 @default.
- W4387378260 cites W2089632738 @default.
- W4387378260 cites W2093534731 @default.
- W4387378260 cites W2097267381 @default.
- W4387378260 cites W2126810579 @default.
- W4387378260 cites W2127501699 @default.
- W4387378260 cites W2150174491 @default.
- W4387378260 cites W2169918686 @default.
- W4387378260 cites W2344975321 @default.
- W4387378260 cites W2388216700 @default.
- W4387378260 cites W2583019817 @default.
- W4387378260 cites W2599251041 @default.
- W4387378260 cites W2654988205 @default.
- W4387378260 cites W2739095506 @default.
- W4387378260 cites W2741907166 @default.
- W4387378260 cites W2756894032 @default.
- W4387378260 cites W2766188050 @default.
- W4387378260 cites W2791548393 @default.
- W4387378260 cites W2898677863 @default.
- W4387378260 cites W2899038052 @default.
- W4387378260 cites W2899283976 @default.
- W4387378260 cites W2899663614 @default.
- W4387378260 cites W2904216680 @default.
- W4387378260 cites W2909386406 @default.
- W4387378260 cites W2912610206 @default.
- W4387378260 cites W2913144232 @default.
- W4387378260 cites W2913994460 @default.
- W4387378260 cites W2926609641 @default.
- W4387378260 cites W2933138175 @default.
- W4387378260 cites W2940585064 @default.
- W4387378260 cites W2952609044 @default.
- W4387378260 cites W2955728019 @default.
- W4387378260 cites W2963341661 @default.
- W4387378260 cites W2963355311 @default.
- W4387378260 cites W2964121744 @default.
- W4387378260 cites W2979896523 @default.
- W4387378260 cites W3016106826 @default.
- W4387378260 cites W3031175989 @default.
- W4387378260 cites W3101658985 @default.
- W4387378260 cites W3102455230 @default.
- W4387378260 cites W3104324110 @default.
- W4387378260 cites W3157504026 @default.
- W4387378260 cites W3160458248 @default.
- W4387378260 cites W3161343553 @default.
- W4387378260 cites W3180220247 @default.
- W4387378260 cites W3198923947 @default.
- W4387378260 cites W3208743843 @default.
- W4387378260 cites W4200300291 @default.
- W4387378260 cites W4212828284 @default.
- W4387378260 cites W4220670607 @default.
- W4387378260 cites W4221074078 @default.
- W4387378260 cites W4224926216 @default.
- W4387378260 cites W4224928099 @default.
- W4387378260 cites W4226498267 @default.
- W4387378260 cites W4281765823 @default.
- W4387378260 cites W4281900408 @default.
- W4387378260 cites W4283332789 @default.
- W4387378260 cites W4285135037 @default.
- W4387378260 cites W4288097622 @default.
- W4387378260 cites W4289550056 @default.
- W4387378260 cites W4293167808 @default.
- W4387378260 cites W4297995647 @default.
- W4387378260 cites W4298177434 @default.
- W4387378260 cites W4300594141 @default.
- W4387378260 cites W4301519015 @default.
- W4387378260 cites W4308343432 @default.
- W4387378260 cites W4308523455 @default.
- W4387378260 cites W4312236358 @default.
- W4387378260 cites W4321764341 @default.
- W4387378260 cites W4323927097 @default.
- W4387378260 cites W4367609927 @default.
- W4387378260 doi "https://doi.org/10.1038/s42256-023-00714-5" @default.
- W4387378260 hasPublicationYear "2023" @default.
- W4387378260 type Work @default.
- W4387378260 citedByCount "0" @default.
- W4387378260 crossrefType "journal-article" @default.
- W4387378260 hasAuthorship W4387378260A5024428666 @default.
- W4387378260 hasAuthorship W4387378260A5048065823 @default.
- W4387378260 hasAuthorship W4387378260A5056821038 @default.