JCLthatlittlerat/Speech-Emotion-Recognition-using-Deep-Learning-for-Amharic-and-Afan-Oromo

This is a speech emotion recognition using deep learning for local languages. The specific languages used in this project is Amharic and Afan oromo which are widely used language in Ethiopia. The Learning method utilized in this projce it Deep Learning and semi-supervided Learning. The dataset are collected from different sources and used as one. # Speech Emotion Recognition using Deep Learning (Amharic and Afan Oromo) Speech emotion recognition using deep learning for local Ethiopian languages. The project focuses on **Amharic** and **Afan Oromo**. The approach combines deep learning and semi-supervised learning. Datasets are collected from multiple sources and combined for training. ## Project layout | File | Description | | --- | --- | | `speech-emotion-recognition-using-deep-learning.ipynb` | Main notebook: log-mel features, CNN–BiLSTM model, training | | `nlp-pipeline-and-dataset-link.txt` | NLP pipeline notes and dataset references | ## Environment setup Use a virtual environment (recommended on Debian/Ubuntu: `sudo apt install python3-venv` if `python3 -m venv` fails): ```bash python3 -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate pip install -r requirements.txt ``` Start Jupyter if you use the notebook locally: ```bash jupyter lab # or: jupyter notebook ``` ## Data The notebook was written for a Kaggle-style path. For local runs, point `data_path` in the notebook to a folder of emotion class subdirectories containing `.wav` files, or use the Amharic Speech Emotion Dataset (ASED) and similar sources described in `nlp-pipeline-and-dataset-link.txt`. ## Contributing 1. Fork the repository and create a branch for your changes. 2. Run the environment setup above and verify the notebook or scripts you touch. 3. Open a pull request with a short description of your changes. ## New contribution: simple project website A basic website is included to make the project easier to understand and contribute to. ### Files added - `index.html` - project landing page - `styles.css` - visual styling - `app.js` - dynamic rendering for pipeline, datasets, and model cards ### How to run website locally 1. Open the project folder. 2. Double-click `index.html` (or run it with a static file server). 3. Use the page sections to review pipeline steps, dataset links, and candidate mod …