This repository provides some data processing pipeline and exploration tools for HTK lattices and confusion networks as required by https://github.com/alecokas/BiLatticeRNN-Confidence.
The scripts are divided into two categories: those which are included in the data processing pipeline and those which are used to obtain statistics or information from the dataset
- python 3.6.3
- pytorch 0.3.1
- numpy 1.14.0
- matplotlib 2.1.2
- scikit-learn 0.19.1
- tqdm 4.19.5