Project Overview
An AI-powered tool designed to transcribe audio, perform Named Entity Recognition (NER), and classify tone from text to support inclusive communication and feedback analysis.
- Audio transcription using OpenAI Whisper
- Named Entity Recognition (NER) with spaCy custom-trained model
- Tone Classification with a custom scikit-learn model
- Confidence score output for model predictions
- Simple web interface or CLI to input audio/text and receive analyzed output
git clone https://github.com/Frostbrewn/ai-inclusion-facilitator.git
cd ai-inclusion-facilitator
python -m venv venv
source venv/bin/activate # macOS/Linux
venv\Scripts\activate # Windows
pip install -r requirements.txtTrain models (optional)
⚠️ Custom train the NER and tone classifier models if needed:
python -m spacy train ner_training/config.cfg --output ner_output --paths.train ner_data/train.spacy --paths.dev ner_data/dev.spacy
python train_tone_classifier.pyRun the app
python run.py