This repository contains the code and results for the ASR project and thesis by Jen Dusseljee.
It contains the following relevant folders and files:
- datasets: The dataset used, including concatenated audio segments.
- jen_original_transcript: Classification on original transcripts.
- jen_whisper: Code for transcription using Whisper and classification on resulting transcripts.
- jen_whisper_medical: Code for transcription using Distil-Whisper and classification on resulting transcripts.
- myprosody_features: Code for acoustic feature extraction and classification.
- jen_roc_curves.ipynb: ROC curves for the thesis.
- prep_data.ipynb: Some code used for dataprocessing.