TTS is a research and production-grade text-to-speech toolkit supporting multiple neural architectures including Glow-TTS, HiFiGAN, MelGAN, Tacotron, Bark, and XTTS. It provides pretrained models in over 1100 languages, tools for training new models, fine-tuning capabilities, and utilities for dataset analysis and curation. The library supports voice cloning, speaker encoding, and streaming inference with low latency.