kvilla/wav2vec-english-speech-emotion-recognition-finetuned

Wav2Vec2ForSpeechClassification

Fine-tuned model based on https://huggingface.co/r-f/wav2vec-english-speech-emotion-recognition. The model is trained on the full TESS, RAVDESS, and CREMA datasets. The Github repository is found here

Authored by:

• —Irene Therese Bermejo Patacsil
• —Kiana Alessandra Villaera

Datasets

• —TESS
• —RAVDESS
• —CREMA-D

Performance Metrics

• —Test accuracy: 85%
• —Test loss: 0.4458

Overview

The model is based on Wav2Vec2 and fine-tuned on the aforementioned datasets.