A system identification approach to determining listening attention from EEG signals

Aličković, Emina; Lunner, Thomas; Gustafsson, Fredrik

doi:10.1109/eusipco.2016.7760204

Cited by 16 publications

(13 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other methods for attention decoding include the forward modelling approach: predicting EEG from the auditory stimulus (Akram et al, 2016;Alickovic et al, 2016), canonical correlation analysis (CCA)-based methods (de Cheveigné et al, 2018), and Bayesian state-space modeling (Miran et al, 2018). The current state-of-the-art models are capable of classifying auditory attention in a two-speaker scenario with high accuracy (80-90% correct) over a data window with a length of approximately 10 s. However, to quickly detect a switch in attention, detection in much shorter windows, down to a few seconds, is required.…”

Section: Introductionmentioning

confidence: 99%

“…The reconstructed envelope is then correlated with the original stimulus envelopes, and the one yielding the highest correlation is then considered to belong to the attended speaker. Other methods for attention decoding include the forward modelling approach: predicting EEG from the auditory stimulus (Akram et al, 2016;Alickovic et al, 2016), canonical correlation analysis (CCA)-based methods (de Cheveigné et al, 2018), and Bayesian state-space modeling (Miran et al, 2018).All studies mentioned above are based on linear decoders. However, since the human auditory system is inherently non-linear (Faure and Korn, 2001), non-linear models (such as neural networks) could be beneficial for reliable and quick AAD.…”

mentioning

confidence: 99%

See 1 more Smart Citation

EEG-based detection of the locus of auditory attention with convolutional neural networks

Vandecappelle

Deckers

Das

et al. 2018

Preprint

View full text Add to dashboard Cite

When multiple people talk simultaneously, the healthy human auditory system is able to attend to one particular speaker of interest. Recently, it has been demonstrated that it is possible to infer to which speaker someone is attending by relating the neural activity, recorded by electroencephalography (EEG), with the speech signals. This is relevant for an effective noise suppression in hearing devices, in order to detect the target speaker in a multi-speaker scenario. Most auditory attention detection algorithms use a linear EEG decoder to reconstruct the attended stimulus envelope, which is then compared to the original stimuli envelopes to determine the attended speaker. Classifying attention within a short time interval remains the main challenge. We present two different convolutional neural network (CNN)-based approaches to solve this problem. One aims to select the attended speaker from a given set of individual speaker envelopes, and the other extracts the locus of auditory attention (left or right), without knowledge of the speech envelopes. Our results show that it is possible to decode attention within 1-2 seconds, with a median accuracy around 80%, without access to the speech envelopes. This is promising for neuro-steered noise suppression in hearing aids, which requires fast and accurate attention detection. Furthermore, the possibility of detecting the locus of auditory attention without access to the speech envelopes is promising for the scenarios in which per-speaker envelopes are unavailable. It will also enable establishing a fast and objective attention measure in future studies. Index TermsConvolutional neural networks (CNN), auditory attention detection (AAD), electroencephalography (EEG), neurosteered auditory prosthesis, brain-computer interface (BCI)

show abstract

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

EEG-based detection of the locus of auditory attention with convolutional neural networks

Vandecappelle

Deckers

Das

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…Furthermore, the robustness of the attended speech envelope reconstruction in noisy real world acoustic scenes has been demonstrated [27]. In contrast to the stimulus reconstruction methods, studies with system identification approaches to solve this problem, have tried to reconstruct the neural measurements using the linear forward map of sound sources [28], [29], [30], [31]. In a recent related study, a single in-Ear-EEG electrode and an adjacent scalp-EEG electrode were used for auditory attention detection in a diotic two speaker scenario [30].…”

Section: Introductionmentioning

confidence: 99%

EEG-assisted modulation of sound sources in the auditory scene

Haghighi

Moghadamfalahi

Akçakaya

et al. 2018

Biomedical Signal Processing and Control

View full text Add to dashboard Cite

Noninvasive EEG (electroencephalography) based auditory attention detection could be useful for improved hearing aids in the future. This work is a novel attempt to investigate the feasibility of online modulation of sound sources by probabilistic detection of auditory attention, using a noninvasive EEG-based brain computer interface. Proposed online system modulates the upcoming sound sources through gain adaptation which employs probabilistic decisions EEG measurements. Second, using the classifier trained offline in the calibration session, we have shown the performance of the online sound source modulation system. We observe that online sound source modulation system is able to keep the level of attended sound source higher than the unattended source.

show abstract

“…This process will be referred to as auditory attention detection (AAD). These models followed mostly two approaches: define a regression model that predicts the neural responses based on the sound, so-called forward modeling [4,2,3], or reconstruct the speech envelopes of the (un)attended speaker from the EEG or MEG neural data (backward modeling) [29,11]. Others have experimented with alternative approaches, for example to extract features from the cross-correlation of neural response and speech envelope to train a linear classifier [19].…”

Section: Introductionmentioning

confidence: 99%

Online detection of auditory attention with mobile EEG: closing the loop with neurofeedback

Zink

Proesmans

Bertrand

et al. 2017

Preprint

View full text Add to dashboard Cite

Auditory attention detection (AAD) is promising for use in auditory-assistive devices to detect to which sound the user is attending. Being able to train subjects in achieving high AAD performance would greatly increase its application potential. In order to do so an acceptable temporal resolution and online implementation are essential prerequisites. Consequently, users of an online AAD can be presented with feedback about their performance. Here we describe two studies that investigate the effects of online AAD with feedback. In the first study, we implemented a fully automated closed-loop system that allows for user-friendly recording environments. Subjects were presented online with visual feedback on their ongoing AAD performance. Following these results we implemented a longitudinal case study in which two subjects were presented with AAD sessions during four weeks. The results prove the feasibility of a fully working online (neuro)feedback system for AAD decoding. The detected changes in AAD for the feedback subject during and after training suggest that changes in AAD may be achieved via training. This is early evidence of such training effects and needs to be confirmed in future studies to evaluate training of AAD in more detail. Finally, the large number of sessions allowed to examine the correlation between the stimuli (i.e. acoustic stories) and AAD performance which was found to be significant. Future studies are suggested to evaluate their acoustic stimuli with care to prevent spurious associations.

show abstract

A system identification approach to determining listening attention from EEG signals

Cited by 16 publications

References 11 publications

EEG-based detection of the locus of auditory attention with convolutional neural networks

EEG-based detection of the locus of auditory attention with convolutional neural networks

EEG-assisted modulation of sound sources in the auditory scene

Online detection of auditory attention with mobile EEG: closing the loop with neurofeedback

Contact Info

Product

Resources

About