Enhancing dysarthric speech recognition through SepFormer and hierarchical attention network models with multistage transfer learning

R Vinotha, D Hepsiba, LD Vijay Anand, J Andrew… - Scientific Reports, 2024 - nature.com
Dysarthria, a motor speech disorder that impacts articulation and speech clarity, presents
significant challenges for Automatic Speech Recognition (ASR) systems. This study …

PSELDNets: Pre-trained Neural Networks on Large-scale Synthetic Datasets for Sound Event Localization and Detection

J Hu, Y Cao, M Wu, F Kang, F Yang, W Wang… - arXiv preprint arXiv …, 2024 - arxiv.org
Sound event localization and detection (SELD) has seen substantial advancements through
learning-based methods. These systems, typically trained from scratch on specific datasets …

Integrating international Chinese visualization teaching and vocational skills training: leveraging attention-connectionist temporal classification models

Y Yao, Z Dai, M Shahbaz - PeerJ Computer Science, 2024 - peerj.com
The teaching of Chinese as a second language has become increasingly crucial for
promoting cross-cultural exchange and mutual learning worldwide. However, traditional …

[PDF][PDF] Towards improved Automatic Speech Recognition for children

T Rolland, A Abad - Transfer, 2024 - isca-archive.org
Abstract Children's Automatic Speech Recognition (ASR) represents a considerable
challenge, with a considerable performance decline of state-of-the-art systems when …

AC-Mix: Self-Supervised Adaptation for Low-Resource Automatic Speech Recognition using Agnostic Contrastive Mixup

C Carvalho, A Abad - arXiv preprint arXiv:2410.14910, 2024 - arxiv.org
Self-supervised learning (SSL) leverages large amounts of unlabelled data to learn rich
speech representations, fostering improvements in automatic speech recognition (ASR) …

Personalized Speech Recognition for Children with Test-Time Adaptation

Z Shi, H Srivastava, X Shi, S Narayanan… - arXiv preprint arXiv …, 2024 - arxiv.org
Accurate automatic speech recognition (ASR) for children is crucial for effective real-time
child-AI interaction, especially in educational applications. However, off-the-shelf ASR …

[PDF][PDF] Accelerat. AI: INESC-ID/IST-Universidade de Lisboa contributions towards improved conversational agents in European Portuguese

A Abad, S Paulo, R Solera-Ureña… - Proc. IberSPEECH …, 2024 - isca-archive.org
Accelerat. AI project aims to create disruptive solutions based on Conversational Artificial
Intelligence (AI) Agents and CCaaS (Contact Center as a Service), which will accelerate …