Audar-TTS-V1

About

Audar-TTS-V1 is an expressive Arabic-priority text-to-speech model with open weights. It produces natural speech directly from raw text for Modern Standard Arabic, various dialects, and Arabic-English code switching, using inline tags such as [laughs] to guide delivery without requiring a phonemizer or G2P system. The model includes a collection of natural synthetic voices. Its Flash and Turbo versions are freely available for download, fine tuning, and self hosting. It is designed for applications in voice agents, media production, education, and accessibility.