Benchmarking Training Paradigms, Dataset Composition, and Model Scaling for Child ASR in ESPnet
Year: 2025
Type: conference-paper
Abstract: sponsorship: FWO-SBO|S004923N: NEFL, KU Leuven|C24M/22/025, FWO|V401325N, National Science Foundation grants|#2138259,#:2138286, #:2138307, #:2137603, and #:2138296
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G1353381076, G1453181110, G1503004319, G1889355044, Track 4: Advanced CI Coordination Ecosystem: Monitoring and Measurement Services, NeLF: Next Level Flemish Speech Recognition, G298330452, G3562386289, Track 1: ACCESS Resource Allocations Marketplace and Platform Services (RAMPS), G4010157413, G4116821888, G4403015908, ACO: An Open CI Ecosystem to Advance Scientific Discovery (OpenCI), Track 3: COre National Ecosystem for CyberinfrasTructure (CONECT), G5367127336, G5464919783, G7291403024, G8108599699, Track 2: Customized Multi-tier Assistance, Training, and Computational Help (MATCH) for End User ACCESS to CI, Collaborative Research: Improving speech technology for better learning outcomes: the case of AAE child speakers
OpenAlex ID: W4414712604