[Windows] and [NVIDIA] Spark-TTS is a high-quality text-to-speech synthesis system that enables voice cloning and speech generation using deep learning models. The system allows users to generate speech with customizable parameters such as pitch, speed, and gender.