Speech:Exps 0300 016

Description
Author: Vitali Taranto

Date: 4-24-2017

Purpose: I named the damn experiment incorrectly. This experiment is to look into VTLN to see if it could be used to improve the model. I am going to run a default experiment, the only difference being that I turn on VTLN. I have read that I need software to estimate the warp parameter during decode, but not train. We will see how this goes.

Details:

Results: