Audio Augmentation for Non-Native Children’s Speech Recognition through Discriminative Learning

Abstract

Automatic speech recognition (ASR) in children is a rapidly evolving field, as children become more accustomed to interacting with virtual assistants, such as Amazon Echo, Cortana, and other smart speakers, and it has advanced the human–computer interaction in recent generations. Furthermore, non-native children are observed to exhibit a diverse range of reading errors during second language (L2) acquisition, such as lexical disfluency, hesitations, intra-word switching, and word repetitions, which are not yet addressed, resulting in ASR’s struggle to recognize non-native children’s speech. The main objective of this study is to develop a non-native children’s speech recognition system on top of feature-space discriminative models, such as feature-space maximum mutual information (fMMI) and boosted feature-space maximum mutual information (fbMMI). Harnessing the collaborative power of speed perturbation-based data augmentation on the original children’s speech corpora yields an effective performance. The corpus focuses on different speaking styles of children, together with read speech and spontaneous speech, in order to investigate the impact of non-native children’s L2 speaking proficiency on speech recognition systems. The experiments revealed that feature-space MMI models with steadily increasing speed perturbation factors outperform traditional ASR baseline models.

Automatic dysarthria detection and severity level assessment using CWT-layered CNN model

“…These models also perform well with end-to-end techniques like CNN models, which improve classification accuracy by directly learning complex representations based on raw waveform data [9,10]. In contrast to traditional methods that depend on intricate feature extraction procedures [11][12][13][14], raw waveform models examine the speech signal directly, making them easier to apply and comprehend [15]. There are several benefits to incorporating CWT-layer in CNN models, one of which is the ability to capture local and global aspects of the signals that are input at various wavelets such as Amor, Morse, and Bump.…”

Closed-set automatic speaker identification using multi-scale recurrent networks in non-native children

“…In fact, social media platforms are the most popular among teens and young adults, with over half of youngsters aged 8-17 are using the internet and maintaining accounts on social media website [5]. However, several speech technology applications face difficulty in identifying non-native children due to various reasons of child's immaturity in the areas of cognition, phonology, and physiological development (vocal tract) [6,7]. In particular, the pitch of children's voice is higher, and perceptual elements like formants appear at higher frequencies [8].…”

Revolutionizing Education: Unlocking the Potential of Asynchronous Video for Interactive Online Learning

“…Traditional synchronous learning models have long been the foundation of educational systems, requiring students to adhere to rigid schedules and geographic constraints (Sinclair et al, 2015). However, these models often fail to cater to the diverse learning styles and busy schedules of today's learners (Radha & Bansal, 2022). Recognizing the need for innovative approaches that embrace flexibility, asynchronous video presents itself as a promising solution (Ringeval et al, 2015).…”