This is the official repository that contains the impementation of an Automatic Speech Recognition system from Nigerian Pidgin to English.
The data containing audio and transcription files can be downloaded as zip file from this drive https://drive.google.com/file/d/1zJvUdZgSUoaH9_QdwN4qwSv3Okr1b1A8/view?usp=sharing
The ASR_Nigeria_Pidgin_Workbook.ipynb explains a step-by-step process for implementation.
Special thanks to our wonderful lecturer Laurent Besacier, amazing colleagues at AMMI and anonymous reviewers for constant support and constructive feedback throughout the project.
This repository borrowed extensively from these repositories - https://github.com/keleog/PidginUNMT https://colab.research.google.com/github/NVIDIA/NeMo/blob/main/tutorials/asr/01_ASR_with_NeMo.ipynb