Developed a real-time text-to-speech feature for the pipecat repository by integrating Sarvam TTS with WebSocket support, enabling live audio generation and interactive streaming capabilities. The implementation focused on updating the Sarvam TTS service to communicate over WebSockets, allowing for interruptible control during speech synthesis. Delivered an example script to demonstrate live interaction, enhancing the user experience for conversational applications. The work utilized Python for both backend logic and API integration, leveraging WebSockets to facilitate low-latency audio streaming. This feature laid the foundation for future streaming-enabled TTS enhancements and addressed the need for responsive, real-time speech synthesis in the platform.