edge-tts

edge-tts gives Python developers Microsoft's neural text-to-speech voices by speaking the same protocol the Edge browser uses for its online Read Aloud service, so no Edge install, no Windows, and no Azure API key are required. The package installs from PyPI with pip or pipx and provides two commands: edge-tts renders text to audio files with optional SRT subtitle output, and edge-playback speaks the synthesis immediately, using the mpv player on non-Windows systems. A large multilingual catalog of Microsoft neural voices is selectable by name, with rate, volume, and pitch adjustable from the command line or the async Python API, and the library serves as the free TTS backend inside many open source reading, dubbing, and assistant projects. The tradeoffs are inherent to the design: synthesis happens on Microsoft's servers, so an internet connection is mandatory, nothing runs offline, and availability depends on an unofficial use of the service rather than a supported API contract. The code is GPL-3.0 licensed, the project holds around 11,600 GitHub stars, and usage itself costs nothing.