Authors: BRTRS-Team

Description: A CNN is used to extract features. Then, the RNN based encoder-decoder model is trained to get the result. Results from the attention based model and the CTC based model are combined to get the final output.