GuessTheMusic: Song Identification from Electroencephalography response

Sonawane, Dhananjay; Miyapuram, Krishna Prasad; Rs, Bharatesh; Lomas, Derek

doi:10.1145/3430984.3431023

Cited by 11 publications

(19 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other studies have also attempted to not include feature extraction steps but the stimuli length in their experiments are significantly smaller than ours (in the order of seconds) [13,15]. In comparison to [16], the EEG responses in NMED-T were to unfamiliar stimuli, which in prior works has shown to be the harder case as classification performance drops when listeners are not familiar to the music stimuli [23].…”

Section: Comparisonsmentioning

confidence: 93%

“…This follows the training approach of the visually evoked EEG response classification models [4]. Other EEG classification studies also do not do across participant generalization because of the high cost of acquiring a large number of new participants when one participant can give you sufficient recordings [14,16].…”

Section: Comparisonsmentioning

confidence: 99%

“…Recent attempts at classifying auditory EEG responses have achieved similar performance ( 84%) all while relying on feature extraction or simple and short acoustic stimuli [14,16]. Study [16] provides a strong example for treating EEG responses as images in classification tasks. They show that when the EEG recordings are treated as 1D time-series in CNN layers, performance stays at chance.…”

Section: Comparisonsmentioning

confidence: 99%

“…Table 3 illustrates a comparative summary across relevant studies consid- ering their best performance and details of the datasets used to achieve that performance. The most comparable study is by Sonawane et al [16] and we focus here on how our approach expands on it. Table 3 shows that our model is trained on EEG responses with the longest stimuli by a large margin on the order of minutes.…”

Section: Comparisonsmentioning

confidence: 99%

“…Yu et al [15] improved the performance to ∼81% by incorporating canonical correlation analysis between DenseNet and pre-trained VGG model that extracted audio features of the experimental stimuli. Most recently, Sonawane et al [16] improved on these approaches and showed that longer and complex stimuli ( 2 mins of music) could be used to evoke EEG responses used as spectral 2D CNN inputs to classify song ID (∼85.0%).…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Image-based eeg classification of brain responses to song recordings

Ramirez-Aristizabal¹,

Ebrahimpour²,

Kello³

2022

Preprint

View full text Add to dashboard Cite

Classifying EEG responses to naturalistic acoustic stimuli is of theoretical and practical importance, but standard approaches are limited by processing individual channels separately on very short sound segments (a few seconds or less). Recent developments have shown classification for music stimuli (∼ 2 mins) by extracting spectral components from EEG and using convolutional neural networks (CNNs). This paper proposes an efficient method to map raw EEG signals to individual songs listened for end-to-end classification. EEG channels are treated as a dimension of a [Channel × Sample] image tile, and images are classified using CNNs. Our experimental results ( 88.7%) compete with state-of-the-art methods (85.0%), yet our classification task is more challenging by processing longer stimuli that were similar to each other in perceptual quality, and were unfamiliar to participants. We also adopt a transfer learning scheme using a pre-trained ResNet-50, confirming the effectiveness of transfer learning despite image domains being unrelated from each other.

show abstract

Section: Comparisonsmentioning

confidence: 93%

Section: Comparisonsmentioning

confidence: 99%