# Darija speech example

Source: the existing Omneity Labs ASAT model page and its `asat-v3.1-voice-demo-20261014-r1/01-male-1.wav` demo asset. This page reuses the same audio bytes. `recorded-result.json` records the exact text, voice, model and SHA256 from that source collection. The saved voice is ASAT-v3.1; the runnable request uses the current public `asat` alias. A new request can produce a different result.

The public model page describes the voice example. This is a text-to-speech sample, not a captured end-to-end conversation. No third-party footage or real customer audio is included. The related GitHub voice starter is separately MIT licensed. No new license grant for the audio is asserted here.
