metadata
title: README
emoji: 🗣
colorFrom: green
colorTo: blue
sdk: static
pinned: false
thumbnail: >-
https://cdn-uploads.huggingface.co/production/uploads/691f625185dae0dcc594795d/X5j-v3uzvdZ1uejooO2XK.png
Kejue AI
Building Arabic-first voice AI.
What We Do
We develop high-quality Arabic text-to-speech (TTS) systems with native support for Gulf and Saudi dialects. Our work spans the full stack — from data collection and speech encoding to model training and low-latency serving.
Focus Areas
- Arabic TTS — Natural, expressive speech synthesis for Modern Standard Arabic and regional dialects
- Voice Cloning — Zero-shot and few-shot voice cloning for Arabic speakers
- Low-Latency Serving — Production-ready inference pipelines optimized for real-time applications
- Speech Data — Large-scale Arabic speech datasets with dialect coverage