none-yet/anime-captions
Viewer β’ Updated β’ 337k β’ 800 β’ 30
How to use DUTAOZHANG/Styele2Code_model2 with pyannote.audio:
from pyannote.audio import Model, Inference
model = Model.from_pretrained("DUTAOZHANG/Styele2Code_model2")
inference = Inference(model)
# inference on the whole file
inference("file.wav")
# inference on an excerpt
from pyannote.core import Segment
excerpt = Segment(start=2.0, end=5.0)
inference.crop("file.wav", excerpt)This repository hosts two separate model weights that need to be used together for the full functionality of Style2Code:
β
Style Encoder Checkpoint (first stage):
Trained via contrastive learning to extract explicit style embeddings.
β
Flan-T5 Generator Checkpoint (second stage):
Fine-tuned to generate code while incorporating the style embeddings.
To use these weights:
1οΈβ£ Clone the Style2Code repository:
git clone https://github.com/zh19980811/Style2Code.git
cd Style2Code
Base model
google/flan-t5-large