TTSFM mirrors OpenAI’s TTS API, providing a compatible interface for converting text to natural-sounding speech based on OpenAI’s GPT-4o mini TTS model. It offers REST and WebSocket endpoints for audio generation, supports 11 voices and 6 audio formats, and includes speed control and long-text handling. The project ships a Python SDK, an interactive web playground for testing, and pre-built Docker images for easy self-hosting.