Year: 2022
Type: conference-paper
Abstract: The objective of this work is to investigate complementary features which can aid the quintessential Mel frequency cepstral coefficients (MFCCs) in the task of closed, limited set word recognition for non-native English speakers of different mother-tongues. Unlike the MFCCs, which are derived from the spectral energy of the speech signal, the proposed frequency-centroids (FCs) encapsulate the spectral centres of the different bands of the speech spectrum, with the bands defined by the Mel filterbank. These features, in combination with the MFCCs, are observed to provide relative performance improvement in English word recognition, particularly under varied noisy conditions. A two-stage Convolution Neural Network (CNN) is used to model the features of the English words uttered with Arabic, French and Spanish accents.
Institutions: Institut National Polytechnique de Toulouse, École Nationale Supérieure d'Électrotechnique, d'Électronique, d'Informatique, d'Hydraulique et des Télécommunications, Indian Institute of Information Technology Dharwad, University College Dublin, Science Foundation Ireland, ADAPT - Centre for Digital Content Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Award: ADAPT_Phase 2
OpenAlex ID: W4286282879