Improved VITS-Based Multilingual AI Speech Synthesis Model with Domain Adaptors and Acoustic Feature Optimization. IJCAI. 2025;49(19). doi:10.31449/inf.v49i19.7622