Lahgtna OmniVoice Demo

State-of-the-art text-to-speech model explicitly trained to support 13 new Arabic Dialects, in addition to 600+ languages!

New Lahgtna Dialects Supported: Egyptian, Saudi, Moroccan, Bahraini, Sudanese, Iraqi, Lebanese, Syrian, Libyan, Palestinian, Tunisian, Algerian, and Yemeni.

  • Voice Clone — Clone any voice from a reference audio
  • Voice Design — Create custom voices with speaker attributes

Powered by Lahgtna OmniVoice (fine-tuned from OmniVoice by Xiaomi AI Lab).

Recommended: 3–10 seconds audio.

Language (optional) / 语种 (可选)

Keep as Auto to auto-detect the language.