A rule-based NLP engine is designed for converting written text to spoken text. This is part of Natural Language Processing course taken by Prof Mausam. Some examples of the text conversion:
- All abbreviations are separated as they are spoken. Example, “U.S.” or “US” is converted to “u s”.
- All dates are converted into words. Example, “29 March 2012” will be converted to “the twenty ninth of march twenty twelve”. “2011-01-25” will be converted to “the twenty fifth of january twenty eleven”.
- All times are converted into words. Example, “04:40 PM” is converted to “four forty p m”. “21:30:12” is converted to “twenty one hours thirty minutes and twelve seconds”
- Currency is also spelled out. “$15.24” is converted to “fifteen dollars and twenty four cents”. “£11” is converted to “eleven pounds”.
Predictions
python run.py --input_path <path_to_input> --solution_path <path_to_solution>Testing
python test.py --ground_truth_path <path_to_ground_truth> --solution_path <path_to_solution>