ddlBoJack/emotion2vec
Self-supervised pre-trained speech representation model for extracting features and training downstream speech emotion recognition classifiers.
Not currently ranked — collecting fresh signals.
star history
Emotion2vec is a self-supervised pre-training framework for speech emotion representation learning, published at ACL 2024. The repository provides PyTorch implementations for extracting features from pre-trained models and training downstream classifiers. It includes the emotion2vec+ foundation models available on ModelScope and Hugging Face, supporting 9-class emotion recognition tasks through iterative fine-tuning.
Frequently asked
- What is ddlBoJack/emotion2vec?
- Self-supervised pre-trained speech representation model for extracting features and training downstream speech emotion recognition classifiers.
- Is emotion2vec open source?
- Yes — ddlBoJack/emotion2vec is an open-source project tracked on heatdrop.
- What language is emotion2vec written in?
- ddlBoJack/emotion2vec is primarily written in Python.
- How popular is emotion2vec?
- ddlBoJack/emotion2vec has 1.1k stars on GitHub.
- Where can I find emotion2vec?
- ddlBoJack/emotion2vec is on GitHub at https://github.com/ddlBoJack/emotion2vec.