Add pipeline tag, library name and link to paper

This PR improves the model card by adding the `library_name`, `pipeline_tag` and link to the paper.
This makes sure that the model is found at https://huggingface.co/models?pipeline_tag=image-to-image as well as links the model to the respective Arxiv page.

README.md

@@ -1,8 +1,12 @@
----
-license: openrail
----
-
-# The checkpoints of LatentSync
-
-This repo not only stores the pretrained U-Net and SyncNet checkpoints of LatentSync, but also stores the whisper checkpoints, auxiliary checkpoints for detecting face, calculating syncnet confidence score and so on. They have covered all you need for both inference and training of LatentSync
-
+---
+license: openrail
+library_name: diffusers
+pipeline_tag: image-to-image
+---
+
+# The checkpoints of LatentSync
+
+This repo not only stores the pretrained U-Net and SyncNet checkpoints of LatentSync, but also stores the whisper checkpoints, auxiliary checkpoints for detecting face, calculating syncnet confidence score and so on. They have covered all you need for both inference and training of LatentSync
+
+The model was presented in the paper [LatentSync: Audio Conditioned Latent Diffusion Models for Lip Sync](https://arxiv.org/abs/2412.09262).
+Code: https://github.com/chunyu-li/LatentSync