Skip to content

Latest commit

 

History

History
20 lines (11 loc) · 771 Bytes

README.md

File metadata and controls

20 lines (11 loc) · 771 Bytes

Korean-Speech-With-Numbers

Pre-processing code for Korean Speech Data with Number provided by AI Hub

Intro

This reposisotry is pre-processing code for Korean Speech Data with Number provided by AI Hub

Korean Speech Data with Number consists of more than 10,000 hours of voice data with 84 categories including numbers in Chinese characters (한자어), native words(고유어) and foreign words(외래어).

Usage

  1. Modify the data_root, data_sets.. options on run.sh

  2. Run 'bash run.sh' to pre-process data

Reference

  1. https://github.com/NVIDIA/NeMo
  2. https://m.blog.naver.com/PostView.naver?isHttpsRedirect=true&blogId=aimldl&logNo=221559323232