so
i'm going to have proposed here with my only collectively from if you k
it's rental
and basically score multichannel i-vector combination for robust speaker verification from the most environments
basically we were well a couple of years ago in a in a project about
a automation application basically was glued inner product and the main characteristics of the project
was uttered below a kind of or
of
speech interface in order to talk to activate the windows the doors and so on
for people with physical impairments
i either they're
them as you mentioned above the main that this is was that the use or
try to users like a multiple microphones are not was from the big microphones with
of the system always listening and of course the use was supposed to be able
to say comments from anywhere in any position in the room
so it we of course that some speaker tasks and some speaker services we were
there's the in and basically the conference we were expecting in this kind of applications
not it was
of course of the speaker can be anyone so we expect to have a huge
mismatch of in between enrollment the spaces
even if you have fixed the problem and or even clean enrolment with a and
i from application about as more from location
and of course we need to the models that were able to cope with this
problem a typically wouldn't five number of microphones the solution that
you can find
in the literature are trying to play something of the speech enhancement even microphone array
beamforming and someone usually you need but it will clearly that devices for that and
sensors you can do some common combination the post processed with the to describe it
some
and to play every with i-vectors or to combine different channels in the sense different
channels that are recorded in a at the same utterance so we have
the same
different samples of the same utterance i for that we would basically use this apartment
with we may sure many impulse responses with a from many locations of positions and
which relate the database
i'm basically we will that the scenario we will focus on some have been a
fixed position for enrollment and you can be anywhere for speaker verification we are explained
every for the combined effect but we can try and we can talk more an
impostor