VoxForge
Hi,
I have installed htk 3.4.1 in window 8. I received the audio files recorded earlier & I'm using Praat software for the phonemes. Now that I want to create speaker independent acoustic model, how can I use files that I have? I do not understand how to do it as the tutorial steps start with prepare data & coding the data.
Do I need to install other things like Cygwin, Julius.
Please help coz I'm really clueless.
Thanks.
--- (Edited on 8/21/2013 8:45 pm [GMT-0500] by newbie) ---
>I want to create speaker independent acoustic model,
Follow the voxforge tutorial to create a speaker dependent acoustic model (using your own speech) to get a basic understanding of the process.
Then update the prompt list from your speaker dependent acoustic model to include prompts recorded by other people, and then retrain the acoustic model - which will start to become a speaker independent acoustic model, becoming more independent as you add more speech.
--- (Edited on 8/21/2013 10:25 pm [GMT-0400] by kmaclean) ---
Thanks, will try it.
--- (Edited on 8/25/2013 11:38 am [GMT-0500] by newbie) ---