# HuBERT fine-tuning

This model is a speech classification model fine-tuned based on facebook/hubert-base-ls960, mainly used for binary classification tasks, and achieved an accuracy of 78.26% on the evaluation set.

A music genre classification model fine-tuned on distilhubert, achieving 84% accuracy on the GTZAN dataset

A Chinese speech emotion classification model fine-tuned on the CASIA dataset using Tencent Game Partner's pre-trained Chinese HuBERT model, supporting 6 emotion categories.

A Thai speech recognition model fine-tuned from facebook/hubert-large-ll60k, trained on data from Common Voice 7.0

Arabic automatic speech recognition model based on HuBERT architecture, fine-tuned on a 300-hour Arabic speech dataset