Information divergence functions play a critical role in statistics and information theory. In this paper we show that a non-parametric f-divergence measure can be used to provide improved bounds on the minimum binary classification probability of error for the case when the training and test data are drawn from the same distribution and for the case where there exists some mismatch between training and test distributions. We confirm the theoretical results by designing feature selection algorithms using the criteria from these bounds and by evaluating the algorithms on a series of pathological speech classification tasks.

These acoustic analyses present a promising tool for rapidly assessing treatment options. Automated measures of baseline speech patterns may enable more selective inclusion criteria and stronger group outcomes within treatment studies.

Neurodegenerative diseases such as amyotrophic lateral sclerosis (ALS) can cause locked-in-syndrome (fully paralyzed but aware). Brain-computer interface (BCI) may be the only option to restore their communication. Current BCIs typically use visual or attention correlates in neural activities to select letters randomly displayed on a screen, which are extremely slow (a few words per minute). Speech-BCIs, which aim to convert the brain activity patterns to speech (neural speech decoding), hold the potential to enable faster communication. Although a few recent studies have shown the potential of neural speech decoding, those are focused on speaker-dependent models. In this study, we investigated speaker-independent neural speech decoding of five continuous phrases from Magnetoencephalography (MEG) signals while 8 subjects produced speech covertly (imagination) or overtly (articulation). We have used both supervised and unsupervised speaker adaptation strategies for implementing a speaker independent model. Experimental results demonstrated that the proposed adaptation-based speakerindependent model has significantly improved decoding performance. To our knowledge, this is the first demonstration of the possibility of speaker-independent neural speech decoding.

Direct decoding of speech from the brain is a faster alternative to current electroencephalography (EEG) speller-based brain-computer interfaces (BCI) in providing communication assistance to locked-in patients. Magnetoencephalography (MEG) has recently shown great potential as a non-invasive neuroimaging modality for neural speech decoding, owing in part to its spatial selectivity over other high-temporal resolution devices. Standard MEG systems have a large number of cryogenically cooled channels/sensors (200 − 300) encapsulated within a fixed liquid helium dewar, precluding their use as wearable BCI devices. Fortunately, recently developed optically pumped magnetometers (OPM) do not require cryogens, and have the potential to be wearable and movable making them more suitable for BCI applications. This design is also modular allowing for customized montages to include only the sensors necessary for a particular task. As the number of sensors bears a heavy influence on the cost, size, and weight of MEG systems, minimizing the number of sensors is critical for designing practical MEG-based BCIs in the future. In this study, we sought to identify an optimal set of MEG channels to decode imagined and spoken phrases from the MEG signals. Using a forward selection algorithm with a support vector machine classifier we found that nine optimally located MEG gradiometers provided higher decoding accuracy compared to using all channels. Additionally, the forward selection algorithm achieved similar performance to dimensionality reduction using a stacked-sparse-autoencoder. Analysis of spatial dynamics of speech decoding suggested that both left and right hemisphere sensors contribute to speech decoding. Sensors approximately located near Broca's area were found to be commonly contributing among the higher-ranked sensors across all subjects. INDEX TERMS autoencoder, brain-computer interface, forward selection algorithm, magnetoencephalography, neural speech decoding, OPM, SVM

No abstract