Fast AI voice synthesis for real-time apps — low-latency text-to-speech with natural-sounding output, built for chatbots, virtual assistants, and live interaction use cases