metadata
license: apache-2.0
language:
- en
tags:
- speech-emotion-recognition
- stress-detection
- wav2vec2
- lstm
- audio
pipeline_tag: audio-classification
Hybrid Wav2Vec2 + LSTM Emotion & Stress Model
This repository contains a trained hybrid Wav2Vec2 + LSTM model for emotion and stress recognition from raw speech audio.
- Input: WAV audio (16 kHz)
- Output: Emotion label + Stress score
Training datasets:
- RAVDESS
- TESS
- SAVEE
Inference code is included.