Bilingual Text-dependent Speaker Verification with Pre-trained Models for TdSV Challenge 2024