We build and maintain models, datasets, and tools for Uzbek voice technology — from speech recognition to text-to-speech.
