Spatial and Spectral Fingerprint in the Brain: Speaker Identification from Single Trial MEG Signals

  title={Spatial and Spectral Fingerprint in the Brain: Speaker Identification from Single Trial MEG Signals},
  author={Debadatta Dash and Paul Ferrari and Jun Wang},
Brain activity signals are unique subject-specific biological features that can not be forged or stolen. Recognizing this inherent trait, brain waves are recently being acknowledged as a far more secure, sensitive, and confidential biometric approach for user identification. Yet, current electroencephalography (EEG) based biometric systems are still in infancy considering their requirement of a large number of sensors and lower recognition performance compared to present biometric modalities… 

Figures from this paper

MEG Sensor Selection for Neural Speech Decoding

It is found that nine optimally located MEG gradiometers provided higher decoding accuracy compared to using all channels, and the forward selection algorithm achieved similar performance to dimensionality reduction using a stacked-sparse-autoencoder.

Brainprints: identifying individuals from magnetoencephalograms

This work proposes three types of interpretable MEG features that can also be used to identify individuals across sessions with high accuracy, and proposes consistent brainprint components that are important for identification.

NeuroVAD: Real-Time Voice Activity Detection from Non-Invasive Neuromagnetic Signals

This study attempted to automatically detect the voice/speech activity directly from the neural signals recorded using magnetoencephalography (MEG), and demonstrated the possibility of real-time VAD direct from the non-invasive neural signals with about 88% accuracy.

Voice of Your Brain: Cognitive Representations of Imagined Speech, Overt Speech, and Speech Perception Based on EEG

The results demonstrate the possibility of subject identification from single channel EEG of imagined speech and overt speech and the comparison of the three speech-related paradigms will provide valuable information for the practical use of speech- related brain signals in the further studies.

Role of Brainwaves in Neural Speech Decoding

Neural speech decoding aims at direct decoding of speech from the brain to restore speech communication in patients with locked-in syndrome (fully paralyzed but aware). Despite the recent progress,

Decoding Speech Evoked Jaw Motion from Non-invasive Neuromagnetic Oscillations

Experimental results indicated that the jaw kinematics can be successfully decoded from non-invasive neural (MEG) signals.

Decoding Imagined and Spoken Phrases From Non-invasive Neural (MEG) Signals

This study investigated the decoding of five imagined and spoken phrases from single-trial, non-invasive magnetoencephalography (MEG) signals collected from eight adult subjects and found convolutional neural networks applied on the spatial, spectral and temporal features extracted from the MEG signals to be highly effective.

Magnetometers vs Gradiometers for Neural Speech Decoding

It is reconfirm that gradiometers are preferable in MEG based decoding analysis but also provides the possibility towards the use of magnetometers (or OPMs) for the development of the next-generation speech-BCIs.

Neural Speech Decoding for Amyotrophic Lateral Sclerosis

Investigation of the decoding of imagined and spoken phrases from non-invasive magnetoencephalography signals of ALS subjects using several spectral features with seven machine learning decoders indicated that the decoding performance for ALS patients is lower than healthy subjects yet significantly higher than chance level.

Neural Speech Decoding with Magnetoencephalography

Promising results have been shown with MEG for speech decoding, providing foundation towards future wearable MEG based speech-BCI applications.



Towards EEG biometrics: pattern matching approaches for user identification

Experimental results show that, the Oz channel provides the best identification accuracy for both ED and DTW methods, and the stimuli of illegal strings and words seem to trigger more distinguishable brain responses.

EEG Based Biometric Framework for Automatic Identity Verification

This study indicates the enormous potential of the EEG biometrics, especially due to its robustness against fraud, and proposes several advances in the feature extraction and classification stages.

Human Brain Distinctiveness Based on EEG Spectral Coherence Connectivity

A novel approach that fuses spectral coherence-based connectivity between different brain regions as a possibly viable biometric feature is proposed and it is suggested that the functional connectivity patterns represent effective features for improving EEG-based biometric systems.

Brain waves for automatic biometric-based user recognition

The issues, which represent an obstacle toward the deployment of biometric systems based on the analysis of brain activity in real life applications are speculated on and a critical and comprehensive review of state-of-the-art methods for electroencephalogram-based automatic user recognition is provided.

Subject identification from electroencephalogram (EEG) signals during imagined speech

  • Katharine BrighamB. Kumar
  • Computer Science
    2010 Fourth IEEE International Conference on Biometrics: Theory, Applications and Systems (BTAS)
  • 2010
The proposed approach was tested on a publicly available database consisting of EEG signals corresponding to Visual Evoked Potentials to test the applicability of the proposed method on a larger number of subjects, and it was able to classify 120 subjects with 98.96% accuracy.

Person Authentication Using Brainwaves (EEG) and Maximum A Posteriori Model Adaptation

  • S. MarcelJ. Millán
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2007
The use of brain activity for person authentication is investigated and a statistical framework based on Gaussian mixture models and maximum a posteriori model adaptation, successfully applied to speaker and face authentication, is proposed, which can deal with only one training session.

Feature selection and channel optimization for biometric identification based on visual evoked potentials

Results in this study revealed the feasibility of VEPs based EEG to be used for biometric identification and the proposed optimization algorithm was shown to have the ability to effectively improve the identification accuracy as well as simplifying the system.

Fisher linear discriminant based person identification using visual evoked potentials

The feasibility of visual evoked potential (VEP) in the gamma band of EEG signal, as a physiological trait, is studied, and used to identify individuals in a group of 20 people.

Novel HHT-Based Features for Biometric Identification Using EEG Signals

  • Su YangF. Deravi
  • Computer Science
    2014 22nd International Conference on Pattern Recognition
  • 2014
In this paper we present a novel approach for biometric identification using electroencephalogram (EEG) signals based on features extracted with the Hilbert-Huang Transform (HHT). The instantaneous

Wavelet-Based EEG Preprocessing for Biometric Applications

  • Su YangF. Deravi
  • Computer Science
    2013 Fourth International Conference on Emerging Security Technologies
  • 2013
This paper compares the effectiveness of different wavelet-based noise removal methods and proposes an EEG-based biometric identification system which combines two such de-noising methods to enhance the signal preprocessing stage.