You can deploy this model and start earning money today!

fwerkor/
content-vec-best

Name: content-vec-best
Availability: InStock

Views

No views yet

Content Vec Best

Official Repo: ContentVec
This repo brings fairseq ContentVec model to HuggingFace Transformers.

How to use

To use this model, you need to define

python

from transformers import HubertModel
import torch.nn as nn
class HubertModelWithFinalProj(HubertModel):
    def __init__(self, config):
        super().__init__(config)

        # The final projection layer is only used for backward compatibility.
        # Following https://github.com/auspicious3000/contentvec/issues/6
        # Remove this layer is necessary to achieve the desired outcome.
        self.final_proj = nn.Linear(config.hidden_size, config.classifier_proj_size)

and then load the model with

python

audio = torch.randn(1, 16000)

model = HubertModelWithFinalProj.from_pretrained("lengyue233/content-vec-best")

x = model(audio)["last_hidden_state"]

How to convert

You need to download the ContentVec_legacy model from the official repo, and then run

python convert.py

fwerkor/content-vec-best

Content Vec Best

How to use

How to convert

fwerkor/
content-vec-best