We want to do recognition for mixed accent of audio samples.
So the nature of audio samples to recognize would be of two people from different region/accents like Indian and Australian.
so our question is how to adapt an Acoustic Model with mixed accents?
Do we need record/prepare the recordings with the two mixed accents or can we record separate samples for two accents and adapt the system?
example: 100 indian and 100 australian recordings where each recording will have only one type of accent (or) Total 200 recording where each sample is mixed with two types of accents.
What would be the best approach?
Last edit: Nagarjuna 2014-12-08
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We want to do recognition for mixed accent of audio samples.
So the nature of audio samples to recognize would be of two people from different region/accents like Indian and Australian.
so our question is how to adapt an Acoustic Model with mixed accents?
Do we need record/prepare the recordings with the two mixed accents or can we record separate samples for two accents and adapt the system?
example: 100 indian and 100 australian recordings where each recording will have only one type of accent (or) Total 200 recording where each sample is mixed with two types of accents.
What would be the best approach?
Last edit: Nagarjuna 2014-12-08
Awaiting for the response. Any help is appreciated.
doesn't really matter