Worked on enhancing Text-to-Speech functionality in the livekit/agents repository, focusing on the Minimax plugin to deliver a richer and more reliable voice interaction experience. Used Python to implement new voice options, improved emotion handling, and robust error logging, addressing both user experience and developer ergonomics. Refactored configuration parameters, including renaming english_normalization to text_normalization and updating the default voice ID, which clarified setup and improved performance. Emphasized backend development and API integration to streamline onboarding and reduce troubleshooting time. The work resulted in a maintainable, extensible TTS feature set that aligns with business goals for improved user and developer outcomes.