Audio and Prompts Discussions

Audio submissions by various dialects
User: anmichael.573
Date: 8/10/2014 9:50 pm
Views: 5345
Rating: 6

Hi All,

I'm new here and I apologize in advance if my question is too amaetuerish. I was trying to adapt my existing language model for British English. I needed data for this and I was wondering if there was any way I could filter out the submissions by accents?


Thanks in advance,


--- (Edited on 8/10/2014 9:50 pm [GMT-0500] by anmichael.573) ---

Re: Audio submissions by various dialects
User: nsh
Date: 8/12/2014 1:52 pm
Views: 2479
Rating: 6

You can write a simple script in your favourite scripting language like Perl or Python which will check Pronunciation dialect field in etc/README file in every archive:

Pronunciation dialect: British English

however, voxforge database doesn't have enough data. It's better to take British english acoustic model from and segment podcasts from BBC with sphinx4, this way you'll get way more British speech data than from Voxforge.


--- (Edited on 8/12/2014 22:52 [GMT+0400] by nsh) ---