So, Basically i had tons of data which word-based dataset. Each of data is absolutely having different length of time.
This is my Approach :
