---
pipeline_tag: text-to-speech
license: mit
---

This repository contains the model as described in [Advancing Zero-shot Text-to-Speech Intelligibility across Diverse Domains via Preference Alignment](https://huggingface.co/papers/2505.04113).

Project Page: https://intalign.github.io/
Github: https://github.com/open-mmlab/Amphion