tae898/erc
The official implementation of "EmoBERTa: Speaker-Aware Emotion Recognition in Conversation with RoBERTa"
Incorporates contextual utterance history (past and future dialogue turns) and speaker identity to improve emotion classification, achieving 65.61% weighted F1 on MELD and 68.57% on IEMOCAP. Provides pre-trained RoBERTa-based models on Hugging Face and a Flask REST API with Docker containerization for deployment, supporting both CPU and GPU inference.
102 stars. No commits in the last 6 months.
Stars
102
Forks
26
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jan 26, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/tae898/erc"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
monologg/GoEmotions-pytorch
Pytorch Implementation of GoEmotions 😍😢😱
gohjiayi/suicidal-text-detection
Building a suicidal text detection model and mental health chatbot with deep learning models and...
monologg/GoEmotions-Korean
Korean version of GoEmotions Dataset 😍😢😱
katha-ai/EmoTx-CVPR2023
[CVPR 2023] Official code repository for "How you feelin'? Learning Emotions and Mental States...
declare-lab/exemplary-empathy
This repository contains the source codes of the paper -- Exemplars-guided Empathetic Response...