Advancing Speech Recognition With No Speech Or With Noisy Speech

Abstract

In this paper we demonstrate end-to-end continuous speech recognition (CSR) using electroencephalography (EEG) signals with no speech signal as input. An attention model based automatic speech recognition (ASR) and connectionist temporal classification (CTC) based ASR systems were implemented for performing recognition. We further demonstrate CSR for noisy speech by fusing with EEG features.

Extended version of our accepted IEEE EUSIPCO 2019 paper with additional results for CTC model based recognition. arXiv admin note: substantial text overlap with arXiv:1906.08045, arXiv:1906.08044