hello good morning

the title of the paper is about from features to speaker vectors by means of

restricted boltzmann machine adaptation

and it proposed an alternative vector based representation for a speakers

based on representation up a broad audience restricted boltzmann machines

and

our preliminary results

a are compatible in a compatible incompatible with the

i-vector approach

for more the channel information that will be available

this in front of the

post or number seven and i'll be happy to see the

thank you