Developed a Text-to-Speech plugin for the livekit/agents repository, enabling seamless voice synthesis within LiveKit Agent workflows through integration with the Fish Audio API. The work focused on establishing a reusable plugin architecture in Python, leveraging asynchronous programming and audio processing techniques to support automated, voice-enabled agent interactions. The implementation included a complete end-to-end integration path, allowing for straightforward activation of text-to-speech features and laying the groundwork for future support of additional voices and languages. Comprehensive documentation was provided to guide usage and scaling, ensuring the plugin could be easily extended and integrated into broader LiveKit Agent solutions.