Simplify architecture description

Set the Model Details architecture field to Qwen3Omni.

README.md

@@ -26,7 +26,7 @@ Proudly built from [iNLP Lab](https://huggingface.co/iNLP-Lab)
 ## Model Details
 * Languages: Hokkien / Minnan (闽南语), primarily Singapore-accent in this release
 * Modalities: Audio in → Audio out: full spoken dialogue, text input/output also supported
-* Architecture: Built on two generations of Qwen-Omni via model editing: Qwen2.5-Omni-3B Thinker + trained Transformer Bridge + Qwen3-Omni Talker
+* Architecture: Qwen3Omni
 * License: