This study will evaluate whether a multilingual, feature-phone interactive voice response (IVR) system with human-verified basal insulin titration improves glycemic control compared with standard clinic-based insulin titration among adults with type 2 diabetes and low health literacy in Pakistan. Approximately 220 participants will be randomly assigned to either the IVR-supported titration strategy or usual clinic-based care. Participants will be followed for 12 weeks, with the primary outcome being change in glycated hemoglobin (HbA1c) from baseline to Week 12. Secondary outcomes include fasting plasma glucose, insulin dose adjustment, hypoglycemia, treatment adherence, patient satisfaction, health literacy-related outcomes, and implementation measures.