#21194Second, we show a significant improvement for speaker adaptation (SA) using the newSI corpus and a small amount of speech from the new (target) speaker.
lr,20-4-H90-1060,ak
word error rate
</term>
on a standard
<term>
grammar
</term>
and
<term>
test set
</term>
from the
<term>
#21149With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standardgrammar and test set from the DARPA Resource Management corpus.
lr,23-6-H90-1060,ak
corpus
</term>
and a small amount of
<term>
speech
</term>
from the
<term>
new ( target ) speaker
#21201Second, we show a significant improvement for speaker adaptation (SA) using the new SI corpus and a small amount ofspeech from the new (target) speaker.
lr,26-3-H90-1060,ak
usual
<term>
pooling
</term>
of all the
<term>
speech data
</term>
from many
<term>
speakers
</term>
prior
#21120In addition, combination of the training speakers is done by averaging the statistics of independently trained models rather than the usual pooling of all thespeech data from many speakers prior to training.
lr,27-2-H90-1060,ak
</term>
, which uses a large amount of
<term>
speech
</term>
from a few
<term>
speakers
</term>
instead
#21075First, we present a new paradigm for speaker-independent (SI) training of hidden Markov models (HMM), which uses a large amount ofspeech from a few speakers instead of the traditional practice of using a little speech from many speakers.
lr-prod,26-4-H90-1060,ak
</term>
and
<term>
test set
</term>
from the
<term>
DARPA Resource Management corpus
</term>
. This performance is comparable
#21155With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar and test set from theDARPA Resource Management corpus.
measure(ment),12-9-H90-1060,ak
</term>
for
<term>
adaptation
</term>
, the
<term>
error rate
</term>
dropped to 4.1 % --- a 45 % reduction
#21258Using only 40 utterances from the target speaker for adaptation, theerror rate dropped to 4.1% --- a 45% reduction in error compared to the SI result.
measure(ment),14-4-H90-1060,ak
recognition
</term>
, we achieved a 7.5 %
<term>
word error rate
</term>
on a standard
<term>
grammar
</term>
#21143With only 12 training speakers for SI recognition, we achieved a 7.5%word error rate on a standard grammar and test set from the DARPA Resource Management corpus.
model,1-7-H90-1060,ak
<term>
new ( target ) speaker
</term>
. A
<term>
probabilistic spectral mapping
</term>
is estimated independently for each
#21211Aprobabilistic spectral mapping is estimated independently for each training (reference) speaker and the target speaker.
model,1-8-H90-1060,ak
the
<term>
target speaker
</term>
. Each
<term>
reference model
</term>
is transformed to the space of the
#21230Eachreference model is transformed to the space of the target speaker and combined by averaging.
model,14-2-H90-1060,ak
speaker-independent ( SI ) training
</term>
of
<term>
hidden Markov models ( HMM )
</term>
, which uses a large amount of
<term>
#21062First, we present a new paradigm for speaker-independent (SI) training ofhidden Markov models ( HMM ), which uses a large amount of speech from a few speakers instead of the traditional practice of using a little speech from many speakers.
model,17-3-H90-1060,ak
statistics of independently trained
<term>
models
</term>
rather than the usual
<term>
pooling
#21111In addition, combination of the training speakers is done by averaging the statistics of independently trainedmodels rather than the usual pooling of all the speech data from many speakers prior to training.
other,10-5-H90-1060,ak
comparable to our best condition for this
<term>
test suite
</term>
, using 109
<term>
training speakers
#21170This performance is comparable to our best condition for thistest suite, using 109 training speakers.
other,10-8-H90-1060,ak
is transformed to the space of the
<term>
target speaker
</term>
and combined by averaging . Using
#21239Each reference model is transformed to the space of thetarget speaker and combined by averaging.
other,15-5-H90-1060,ak
<term>
test suite
</term>
, using 109
<term>
training speakers
</term>
. Second , we show a significant
#21175This performance is comparable to our best condition for this test suite, using 109training speakers.
other,16-7-H90-1060,ak
reference ) speaker
</term>
and the
<term>
target speaker
</term>
. Each
<term>
reference model
</term>
#21226A probabilistic spectral mapping is estimated independently for each training (reference) speaker and thetarget speaker.
other,22-4-H90-1060,ak
a standard
<term>
grammar
</term>
and
<term>
test set
</term>
from the
<term>
DARPA Resource Management
#21151With only 12 training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar andtest set from the DARPA Resource Management corpus.
other,26-6-H90-1060,ak
amount of
<term>
speech
</term>
from the
<term>
new ( target ) speaker
</term>
. A
<term>
probabilistic spectral mapping
#21204Second, we show a significant improvement for speaker adaptation (SA) using the new SI corpus and a small amount of speech from thenew ( target ) speaker.
other,3-4-H90-1060,ak
<term>
training
</term>
. With only 12
<term>
training speakers
</term>
for
<term>
SI recognition
</term>
, we
#21132With only 12training speakers for SI recognition, we achieved a 7.5% word error rate on a standard grammar and test set from the DARPA Resource Management corpus.
other,3-9-H90-1060,ak
combined by averaging . Using only 40
<term>
utterances
</term>
from the
<term>
target speaker
</term>
#21249Using only 40utterances from the target speaker for adaptation, the error rate dropped to 4.1% --- a 45% reduction in error compared to the SI result.