ViTextnormASR

Step 1:

Create a folder named dataset

Step 2:

Contact us to get our full dataset or try our sample dataset from directory samples and extract them to dataset

Step 3:

Run all the following bash files:

• ./run_mbert.sh
• ./run_vibert.sh
• ./run_velectra.sh

Reference:

@{vitextnormasr,
title     = {{Transformer-based Joint Learning Approach for Text Normalization in Vietnamese ASR}},
author    = {Viet The Bui, Tho Chi Luong, Oanh Thi Tran},
year      = {2022}
}