Correlational Networks for Speaker Normalization in Automatic Speech Recognition

Sharon, Rini A.; Kothinti, Sandeep; Umesh, S.

doi:10.21437/interspeech.2018-1612

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Improving single mode testing performance by using data from other modes during training is one of the applications that motivate the importance of CRL [29]. Correlation Network(CorrNet) based CRL approaches achieve the same [30,31]. CorrNets are trained to implicitly maximize the correlation between the views in a common projected space while maintaining balanced self and cross reconstruction accuracies within the views.…”

Section: Motivation and Related Workmentioning

confidence: 99%

Correlation based Multi-phasal models for improved imagined speech EEG recognition

Sharon¹,

Murthy²

2020

Workshop on Speech, Music and Mind (SMM 2020)

Self Cite

View full text Add to dashboard Cite

Translation of imagined speech electroencephalogram(EEG) into human understandable commands greatly facilitates the design of naturalistic brain computer interfaces. To achieve improved imagined speech unit classification, this work aims to profit from the parallel information contained in multi-phasal EEG data recorded while speaking, imagining and performing articulatory movements corresponding to specific speech units. A bi-phase common representation learning module using neural networks is designed to model the correlation and reproducibility between an analysis phase and a support phase. The trained Correlation Network is then employed to extract discriminative features of the analysis phase. These features are further classified into five binary phonological categories using machine learning models such as Gaussian mixture based hidden Markov model and deep neural networks. The proposed approach further handles the non-availability of multi-phasal data during decoding. Topographic visualizations along with result-based inferences suggest that the multi-phasal correlation modelling approach proposed in the paper enhances imaginedspeech EEG recognition performance.

show abstract

Section: Motivation and Related Workmentioning

confidence: 99%