Publications by Joe Frankel
joe.bib
@inproceedings{frankel07:AF_MLP,
author = {Frankel, J. and Magimai-Doss, M. and King, S. and
Livescu, K. and Çetin, Ö.},
title = {Articulatory Feature Classifiers Trained on 2000 hours
of Telephone Speech},
booktitle = {Proc. Interspeech},
address = {Antwerp, Belgium},
abstract = {This paper is intended to advertise the public
availability of the articulatory feature (AF)
classification multi-layer perceptrons (MLPs) which
were used in the Johns Hopkins 2006 summer workshop. We
describe the design choices, data preparation, AF label
generation, and the training of MLPs for feature
classification on close to 2000 hours of telephone
speech. In addition, we present some analysis of the
MLPs in terms of classification accuracy and confusions
along with a brief summary of the results obtained
during the workshop using the MLPs. We invite
interested parties to make use of these MLPs.},
month = aug,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/frankel_AF-MLP.pdf},
year = 2007
}
@inproceedings{vipperla08,
author = {Ravichander Vipperla and Steve Renals and Joe Frankel},
title = {Longitudinal study of {ASR} performance on ageing
voices},
booktitle = {Proc.~Interspeech},
address = {Brisbane},
abstract = {This paper presents the results of a longitudinal
study of ASR performance on ageing voices. Experiments
were conducted on the audio recordings of the
proceedings of the Supreme Court Of The United States
(SCOTUS). Results show that the Automatic Speech
Recognition (ASR) Word Error Rates (WERs) for elderly
voices are significantly higher than those of adult
voices. The word error rate increases gradually as the
age of the elderly speakers increase. Use of maximum
likelihood linear regression (MLLR) based speaker
adaptation on ageing voices improves the WER though the
performance is still considerably lower compared to
adult voices. Speaker adaptation however reduces the
increase in WER with age during old age.},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2008/vipperla_is08.pdf},
year = 2008
}
@article{Wang_JCST2012,
author = {Dong Wang and Javier Tejedor and Simon King and Joe
Frankel},
title = {Term-dependent Confidence Normalization for
Out-of-Vocabulary Spoken Term Detection},
journal = {Journal of Computer Science and Technology},
volume = {27},
number = {2},
abstract = {Spoken Term Detection (STD) is a fundamental component
of spoken information retrieval systems. A key task of
an STD system is to determine reliable detections and
reject false alarms based on certain confidence
measures. The detection posterior probability, which is
often computed from lattices, is a widely used
confidence measure. However, a potential problem of
this confidence measure is that the confidence scores
of detections of all search terms are treated
uniformly, regardless of how much they may differ in
terms of phonetic or linguistic properties. This
problem is particularly evident for out-of-vocabulary
(OOV) terms which tend to exhibit high intra-term
diversity. To address the discrepancy on confidence
levels that the same confidence score may convey for
different terms, a term-dependent decision strategy is
desirable – for example, the term-specific threshold
(TST) approach. In this work, we propose a
term-dependent normalisation technique which
compensates for term diversity on confidence
estimation. Particularly, we propose a linear bias
compensation and a discriminative compensation to deal
with the bias problem that is inherent in lattice-based
confidence measuring from which the TST approach
suffers. We tested the proposed technique on speech
data from the multi-party meeting domain with two
state-of-the-art STD systems based on phonemes and
words respectively. The experimental results
demonstrate that the confidence normalisation approach
leads to a significant performance improvement in STD,
particularly for OOV terms with phoneme-based systems.},
doi = {http://dx.doi.org/10.1007/s11390-012-1228-x},
year = 2012
}
@inproceedings{wang:frankel:tejedor:king:icassp2008,
author = {Dong Wang and Joe Frankel and Javier Tejedor and Simon
King},
title = {A comparison of phone and grapheme-based spoken term
detection},
booktitle = {Proc. ICASSP},
pages = {4969--4972 },
abstract = {We propose grapheme-based sub-word units for spoken
term detection (STD). Compared to phones, graphemes
have a number of potential advantages. For
out-of-vocabulary search terms, phone- based approaches
must generate a pronunciation using letter-to-sound
rules. Using graphemes obviates this potentially
error-prone hard decision, shifting pronunciation
modelling into the statistical models describing the
observation space. In addition, long-span grapheme
language models can be trained directly from large text
corpora. We present experiments on Spanish and English
data, comparing phone and grapheme-based STD. For
Spanish, where phone and grapheme-based systems give
similar transcription word error rates (WERs),
grapheme-based STD significantly outperforms a phone-
based approach. The converse is found for English,
where the phone-based system outperforms a grapheme
approach. However, we present additional analysis which
suggests that phone-based STD performance levels may be
achieved by a grapheme-based approach despite lower
transcription accuracy, and that the two approaches may
usefully be combined. We propose a number of directions
for future development of these ideas, and suggest that
if grapheme-based STD can match phone-based
performance, the inherent flexibility in dealing with
out-of-vocabulary terms makes this a desirable
approach.},
doi = {10.1109/ICASSP.2008.4518773},
month = {March-April},
year = 2008
}
@article{frankel07:factoring,
author = {Frankel, J. and King, S.},
title = {Factoring {G}aussian Precision Matrices for Linear
Dynamic Models},
journal = {Pattern Recognition Letters},
volume = {28},
number = {16},
pages = {2264-2272},
abstract = {The linear dynamic model (LDM), also known as the
Kalman filter model, has been the subject of research
in the engineering, control, and more recently, machine
learning and speech technology communities. The
Gaussian noise processes are usually assumed to have
diagonal, or occasionally full, covariance matrices. A
number of recent papers have considered modelling the
precision rather than covariance matrix of a Gaussian
distribution, and this work applies such ideas to the
LDM. A Gaussian precision matrix P can be factored into
the form P = UTSU where U is a transform and S a
diagonal matrix. By varying the form of U, the
covariance can be specified as being diagonal or full,
or used to model a given set of spatial dependencies.
Furthermore, the transform and scaling components can
be shared between models, allowing richer distributions
with only marginally more parameters than required to
specify diagonal covariances. The method described in
this paper allows the construction of models with an
appropriate number of parameters for the amount of
available training data. We provide illustrative
experimental results on synthetic and real speech data
in which models with factored precision matrices and
automatically-selected numbers of parameters are as
good as or better than models with diagonal covariances
on small data sets and as good as models with full
covariance matrices on larger data sets.},
categories = {LDM},
doi = {10.1016/j.patrec.2007.07.008},
month = {December},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Frankel_LDM_covar.pdf},
year = 2007
}
@inproceedings{cetin07:crosslingual,
author = {Çetin, Ö. and Magimai-Doss, M. and Kantor, A. and
King, S. and Bartels, C. and Frankel, J. and Livescu,
K.},
title = {Monolingual and crosslingual comparison of tandem
features derived from articulatory and phone {MLP}s},
booktitle = {Proc. ASRU},
address = {Kyoto},
organization = {IEEE},
abstract = {In recent years, the features derived from posteriors
of a multilayer perceptron (MLP), known as tandem
features, have proven to be very effective for
automatic speech recognition. Most tandem features to
date have relied on MLPs trained for phone
classification. We recently showed on a relatively
small data set that MLPs trained for articulatory
feature classification can be equally effective. In
this paper, we provide a similar comparison using MLPs
trained on a much larger data set - 2000 hours of
English conversational telephone speech. We also
explore how portable phone- and articulatory feature-
based tandem features are in an entirely different
language - Mandarin - without any retraining. We find
that while phone-based features perform slightly better
in the matched-language condition, they perform
significantly better in the cross-language condition.
Yet, in the cross-language condition, neither approach
is as effective as the tandem features extracted from
an MLP trained on a relatively small amount of
in-domain data. Beyond feature concatenation, we also
explore novel observation modelling schemes that allow
for greater flexibility in combining the tandem and
standard features at hidden Markov model (HMM) outputs.},
month = {December},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Cetin_etal_ASRU2007.pdf},
year = 2007
}
@inproceedings{frankel01:alternative,
author = {Frankel, J. and King, S.},
title = {Speech recognition in the articulatory domain:
investigating an alternative to acoustic {HMM}s},
booktitle = {Proc. Workshop on Innovations in Speech Processing},
abstract = {We describe a speech recognition system which uses a
combination of acoustic and articulatory features as
input. Linear dynamic models capture the trajectories
which characterize each segment type. We describe
classification and recognition tasks for systems based
on acoustic data in conjunction with both real and
automatically recovered articulatory parameters.},
categories = {am,artic,asr,ldm,mocha,edinburgh},
month = apr,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2001/Frankel_King_WISP2001.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2001/Frankel_King_WISP2001.ps},
year = 2001
}
@article{5510125,
author = {Wang, D. and King, S. and Frankel, J.},
title = {Stochastic Pronunciation Modelling for
Out-of-Vocabulary Spoken Term Detection},
journal = {Audio, Speech, and Language Processing, IEEE
Transactions on},
volume = {PP},
number = {99},
abstract = {Spoken term detection (STD) is the name given to the
task of searching large amounts of audio for
occurrences of spoken terms, which are typically single
words or short phrases. One reason that STD is a hard
task is that search terms tend to contain a
disproportionate number of out-of-vocabulary (OOV)
words. The most common approach to STD uses subword
units. This, in conjunction with some method for
predicting pronunciations of OOVs from their written
form, enables the detection of OOV terms but
performance is considerably worse than for
in-vocabulary terms. This performance differential can
be largely attributed to the special properties of
OOVs. One such property is the high degree of
uncertainty in the pronunciation of OOVs. We present a
stochastic pronunciation model (SPM) which explicitly
deals with this uncertainty. The key insight is to
search for all possible pronunciations when detecting
an OOV term, explicitly capturing the uncertainty in
pronunciation. This requires a probabilistic model of
pronunciation, able to estimate a distribution over all
possible pronunciations. We use a joint-multigram model
(JMM) for this and compare the JMM-based SPM with the
conventional soft match approach. Experiments using
speech from the meetings domain demonstrate that the
SPM performs better than soft match in most operating
regions, especially at low false alarm probabilities.
Furthermore, SPM and soft match are found to be
complementary: their combination provides further
performance gains.},
categories = {confidence estimation, spoken term detection, speech
recognition, OOVs},
doi = {10.1109/TASL.2010.2058800},
issn = {1558-7916},
month = jul,
year = 2010
}
@article{frankel07:AF_DBN,
author = {Frankel, J. and Wester, M. and King, S.},
title = {Articulatory feature recognition using dynamic
{B}ayesian networks},
journal = {Computer Speech & Language },
volume = {21},
number = {4},
pages = {620--640},
abstract = {We describe a dynamic Bayesian network for
articulatory feature recognition. The model is intended
to be a component of a speech recognizer that avoids
the problems of conventional ``beads-on-a-string''
phoneme-based models. We demonstrate that the model
gives superior recognition of articulatory features
from the speech signal compared with a stateof- the art
neural network system. We also introduce a training
algorithm that offers two major advances: it does not
require time-aligned feature labels and it allows the
model to learn a set of asynchronous feature changes in
a data-driven manner.},
month = oct,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Frankel_etal_CSL2007.pdf},
year = 2007
}
@inproceedings{dongwang_interspeech09_spm,
author = {Dong Wang and Simon King and Joe Frankel},
title = {Stochastic Pronunciation Modelling for Spoken Term
Detection},
booktitle = {Proc. of Interspeech},
pages = {2135--2138},
address = {Brighton, UK},
abstract = {A major challenge faced by a spoken term detection
(STD) system is the detection of out-of-vocabulary
(OOV) terms. Although a subword-based STD system is
able to detect OOV terms, performance reduction is
always observed compared to in-vocabulary terms.
Current approaches to STD do not acknowledge the
particular properties of OOV terms, such as
pronunciation uncertainty. In this paper, we use a
stochastic pronunciation model to deal with the
uncertain pronunciations of OOV terms. By considering
all possible term pronunciations, predicted by a
joint-multigram model, we observe a significant
performance improvement. },
categories = {joint-multigram, pronunciation model, spoken term
detection, speech recognition},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2009/spm.pdf},
year = 2009
}
@inproceedings{janin06:rt06s,
author = {Janin, A. and Stolcke, A. and Anguera, X. and Boakye,
K. and Çetin, Ö. and Frankel, J. and Zheng, J.},
title = {The {ICSI-SRI} Spring 2006 Meeting Recognition System},
booktitle = {Proc. MLMI},
address = {Washington DC.},
abstract = {We describe the development of the ICSI-SRI speech
recognition system for the National Institute of
Standards and Technology (NIST) Spring 2006 Meeting
Rich Transcription (RT-06S) evaluation, highlighting
improvements made since last year, including
improvements to the delay-and-sum algorithm, the
nearfield segmenter, language models, posterior-based
features, HMM adaptation methods, and adapting to a
small amount of new lecture data. Results are reported
on RT-05S and RT-06S meeting data. Compared to the
RT-05S conference system, we achieved an overall
improvement of 4\% relative in the MDM and SDM
conditions, and 11\% relative in the IHM condition. On
lecture data, we achieved an overall improvement of 8\%
relative in the SDM condition, 12\% on MDM, 14\% on
ADM, and 15\% on IHM.},
categories = {am,asr},
month = {May},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2006/Janin_et_al_RT06s.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2006/Janin_et_al_RT06s.ps},
year = 2006
}
@inproceedings{Cetin07:tandem,
author = {Çetin, Ö. and Kantor, A. and King, S. and Bartels,
C. and Magimai-Doss, M. and Frankel, J. and Livescu, K.},
title = {An articulatory feature-based tandem approach and
factored observation modeling},
booktitle = {Proc. ICASSP},
address = {Honolulu},
abstract = {The so-called tandem approach, where the posteriors of
a multilayer perceptron (MLP) classifier are used as
features in an automatic speech recognition (ASR)
system has proven to be a very effective method. Most
tandem approaches up to date have relied on MLPs
trained for phone classification, and appended the
posterior features to some standard feature hidden
Markov model (HMM). In this paper, we develop an
alternative tandem approach based on MLPs trained for
articulatory feature (AF) classification. We also
develop a factored observation model for characterizing
the posterior and standard features at the HMM outputs,
allowing for separate hidden mixture and state-tying
structures for each factor. In experiments on a subset
of Switchboard, we show that the AFbased tandem
approach is as effective as the phone-based approach,
and that the factored observation model significantly
outperforms the simple feature concatenation approach
while using fewer parameters.},
month = {April},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Cetin_icassp07_tandem.pdf},
year = 2007
}
@inproceedings{livescu07:JHU_summary,
author = {Livescu, K. and Çetin, Ö. and Hasegawa-Johnson, M.
and King, S. and Bartels, C. and Borges, N. and Kantor,
A. and Lal, P. and Yung, L. and Bezman,
Dawson-Haggerty, S. and Woods, B. and Frankel, J. and
Magimai-Doss, M. and Saenko, K.},
title = {Articulatory feature-based methods for acoustic and
audio-visual speech recognition: {S}ummary from the
2006 {JHU} {S}ummer {W}orkshop},
booktitle = {Proc. ICASSP},
address = {Honolulu},
abstract = {We report on investigations, conducted at the 2006
Johns HopkinsWorkshop, into the use of articulatory
features (AFs) for observation and pronunciation models
in speech recognition. In the area of observation
modeling, we use the outputs of AF classiers both
directly, in an extension of hybrid HMM/neural network
models, and as part of the observation vector, an
extension of the tandem approach. In the area of
pronunciation modeling, we investigate a model having
multiple streams of AF states with soft synchrony
constraints, for both audio-only and audio-visual
recognition. The models are implemented as dynamic
Bayesian networks, and tested on tasks from the
Small-Vocabulary Switchboard (SVitchboard) corpus and
the CUAVE audio-visual digits corpus. Finally, we
analyze AF classication and forced alignment using a
newly collected set of feature-level manual
transcriptions.},
month = {April},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/livescu_icassp07_sum.pdf},
year = 2007
}
@inproceedings{wester04:asynch,
author = {Wester, M. and Frankel, J. and King, S.},
title = {Asynchronous Articulatory Feature Recognition Using
Dynamic {B}ayesian Networks},
booktitle = {Proc. IEICI Beyond HMM Workshop},
address = {Kyoto},
abstract = {This paper builds on previous work where dynamic
Bayesian networks (DBN) were proposed as a model for
articulatory feature recognition. Using DBNs makes it
possible to model the dependencies between features, an
addition to previous approaches which was found to
improve feature recognition performance. The DBN
results were promising, giving close to the accuracy of
artificial neural nets (ANNs). However, the system was
trained on canonical labels, leading to an overly
strong set of constraints on feature co-occurrence. In
this study, we describe an embedded training scheme
which learns a set of data-driven asynchronous feature
changes where supported in the data. Using a subset of
the OGI Numbers corpus, we describe articulatory
feature recognition experiments using both
canonically-trained and asynchronous DBNs. Performance
using DBNs is found to exceed that of ANNs trained on
an identical task, giving a higher recognition
accuracy. Furthermore, inter-feature dependencies
result in a more structured model, giving rise to fewer
feature combinations in the recognition output. In
addition to an empirical evaluation of this modelling
approach, we give a qualitative analysis, comparing
asynchrony found through our data-driven methods to the
asynchrony which may be expected on the basis of
linguistic knowledge.},
categories = {am,artic,asr,dbn,oginumbers,edinburgh},
month = dec,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2004/Wester_et_al_IEICE.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2004/Wester_et_al_IEICE.ps},
year = 2004
}
@inproceedings{dongwang_icassp09,
author = {Dong Wang and Tejedor Tejedor and Joe Frankel and
Simon King},
title = {Posterior-based confidence measures for spoken term
detection},
booktitle = {Proc. of ICASSP09},
address = {Taiwan},
abstract = {Confidence measures play a key role in spoken term
detection (STD) tasks. The confidence measure expresses
the posterior probability of the search term appearing
in the detection period, given the speech. Traditional
approaches are based on the acoustic and language model
scores for candidate detections found using automatic
speech recognition, with Bayes' rule being used to
compute the desired posterior probability. In this
paper, we present a novel direct posterior-based
confidence measure which, instead of resorting to the
Bayesian formula, calculates posterior probabilities
from a multi-layer perceptron (MLP) directly. Compared
with traditional Bayesian-based methods, the
direct-posterior approach is conceptually and
mathematically simpler. Moreover, the MLP-based model
does not require assumptions to be made about the
acoustic features such as their statistical
distribution and the independence of static and dynamic
co-efficients. Our experimental results in both English
and Spanish demonstrate that the proposed direct
posterior-based confidence improves STD performance. },
categories = {Spoken term detection, confidence measure, posterior
probabilities, MLP},
month = {April},
page = {4889--4892},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2009/posterior.pdf},
year = 2009
}
@inproceedings{frankel01:ASR,
author = {Frankel, J. and King, S.},
title = {{ASR} - Articulatory Speech Recognition},
booktitle = {Proc. {E}urospeech},
pages = {599-602},
address = {Aalborg, Denmark},
abstract = {In this paper we report recent work on a speech
recognition system using a combination of acoustic and
articulatory features as input. Linear dynamic models
are used to capture the trajectories which characterize
each segment type. We describe classification and
recognition tasks for systems based on acoustic data in
conjunction with both real and automatically recovered
articulatory parameters.},
categories = {am,artic,asr,ldm,mocha,edinburgh},
month = sep,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2001/Frankel_King_Eurospeech2001.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2001/Frankel_King_Eurospeech2001.ps},
year = 2001
}
@inproceedings{frankel04:artic_dbn,
author = {Frankel, J. and Wester, M. and King, S.},
title = {Articulatory feature recognition using dynamic
{B}ayesian networks},
booktitle = {Proc. {ICSLP}},
abstract = {This paper describes the use of dynamic Bayesian
networks for the task of articulatory feature
recognition. We show that by modeling the dependencies
between a set of 6 multi-leveled articulatory features,
recognition accuracy is increased over an equivalent
system in which features are considered independent.
Results are compared to those found using artificial
neural networks on an identical task.},
categories = {am,artic,asr,dbn,timit,edinburgh},
month = sep,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2004/Frankel_et_al_ICSLP2004.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2004/Frankel_et_al_ICSLP2004.ps},
year = 2004
}
@inproceedings{dongwang_interspeech09_conf,
author = {Dong Wang and Simon King and Joe Frankel and Peter
Bell},
title = {Term-Dependent Confidence for Out-of-Vocabulary Term
Detection},
booktitle = {Proc. Interspeech},
pages = {2139--2142},
address = {Brighton, UK},
abstract = { Within a spoken term detection (STD) system, the
decision maker plays an important role in retrieving
reliable detections. Most of the state-of-the-art STD
systems make decisions based on a confidence measure
that is term-independent, which poses a serious problem
for out-of-vocabulary (OOV) term detection. In this
paper, we study a term-dependent confidence measure
based on confidence normalisation and discriminative
modelling, particularly focusing on its remarkable
effectiveness for detecting OOV terms. Experimental
results indicate that the term-dependent confidence
provides much more significant improvement for OOV
terms than terms in-vocabulary. },
categories = {joint-multigram, pronunciation model, spoken term
detection, speech recognition},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2009/conf.pdf},
year = 2009
}
@inproceedings{vipperla2010a,
author = {Vipperla, Ravi Chander and Renals, Steve and Frankel,
Joe},
title = {Augmentation of adaptation data},
booktitle = {Proc. Interspeech},
pages = {530--533},
address = {Makuhari, Japan},
abstract = {Linear regression based speaker adaptation approaches
can improve Automatic Speech Recognition (ASR) accuracy
significantly for a target speaker. However, when the
available adaptation data is limited to a few seconds,
the accuracy of the speaker adapted models is often
worse compared with speaker independent models. In this
paper, we propose an approach to select a set of
reference speakers acoustically close to the target
speaker whose data can be used to augment the
adaptation data. To determine the acoustic similarity
of two speakers, we propose a distance metric based on
transforming sample points in the acoustic space with
the regression matrices of the two speakers. We show
the validity of this approach through a speaker
identification task. ASR results on SCOTUS and AMI
corpora with limited adaptation data of 10 to 15
seconds augmented by data from selected reference
speakers show a significant improvement in Word Error
Rate over speaker independent and speaker adapted
models.},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2010/vipperla-is2010.pdf},
year = 2010
}
@inproceedings{dong_ivan_joe_simon_interspeech08_marray,
author = {Dong Wang and Ivan Himawan and Joe Frankel and Simon
King},
title = {A Posterior Approach for Microphone Array Based Speech
Recognition},
booktitle = {Proc. Interspeech},
pages = {996--999},
abstract = {Automatic speech recognition (ASR) becomes rather
difficult in meetings domains because of the adverse
acoustic conditions, including more background noise,
more echo and reverberation and frequent cross-talking.
Microphone arrays have been demonstrated able to boost
ASR performance dramatically in such noisy and
reverberant environments, with various beamforming
algorithms. However, almost all existing beamforming
measures work in the acoustic domain, resorting to
signal processing theories and geometric explanation.
This limits their application, and induces significant
performance degradation when the geometric property is
unavailable or hard to estimate, or if heterogenous
channels exist in the audio system. In this paper, we
preset a new posterior-based approach for array-based
speech recognition. The main idea is, instead of
enhancing speech signals, we try to enhance the
posterior probabilities that frames belonging to
recognition units, e.g., phones. These enhanced
posteriors are then transferred to posterior
probability based features and are modeled by HMMs,
leading to a tandem ANN-HMM hybrid system presented by
Hermansky et al.. Experimental results demonstrated the
validity of this posterior approach. With the posterior
accumulation or enhancement, significant improvement
was achieved over the single channel baseline.
Moreover, we can combine the acoustic enhancement and
posterior enhancement together, leading to a hybrid
acoustic-posterior beamforming approach, which works
significantly better than just the acoustic
beamforming, especially in the scenario with
moving-speakers. },
categories = {speech recognition, microphone array, beamforming,
tandem approach},
month = sep,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2008/marray.a.pdf},
year = 2008
}
@article{frankel06:adapt,
author = {Frankel, J. and King, S.},
title = {Observation Process Adaptation for Linear Dynamic
Models},
journal = {Speech Communication},
volume = 48,
number = 9,
pages = {1192-1199},
abstract = {This work introduces two methods for adapting the
observation process parameters of linear dynamic models
(LDM) or other linear-Gaussian models. The first method
uses the expectation-maximization (EM) algorithm to
estimate transforms for location and covariance
parameters, and the second uses a generalized EM (GEM)
approach which reduces computation in making updates
from $O(p^6)$ to $O(p^3)$, where $p$ is the feature
dimension. We present the results of speaker adaptation
on TIMIT phone classification and recognition
experiments with relative error reductions of up to
$6\%$. Importantly, we find minimal differences in the
results from EM and GEM. We therefore propose that the
GEM approach be applied to adaptation of hidden Markov
models which use non-diagonal covariances. We provide
the necessary update equations.},
categories = {am,asr,ldm,timit,edinburgh},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2006/Frankel_King_SPECOM2006.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2006/Frankel_King_SPECOM2006.ps},
year = 2006
}
@article{vipperla2010,
author = {Vipperla, Ravi Chander and Renals, Steve and Frankel,
Joe},
title = {Ageing voices: The effect of changes in voice
parameters on {ASR} performance},
journal = {EURASIP Journal on Audio, Speech, and Music Processing},
abstract = {With ageing, human voices undergo several changes
which are typically characterized by increased
hoarseness and changes in articulation patterns. In
this study, we have examined the effect on Automatic
Speech Recognition (ASR) and found that the Word Error
Rates (WER) on older voices is about 9\% absolute
higher compared to those of adult voices. Subsequently,
we compared several voice source parameters including
fundamental frequency, jitter, shimmer, harmonicity and
cepstral peak prominence of adult and older males.
Several of these parameters show statistically
significant difference for the two groups. However,
artificially increasing jitter and shimmer measures do
not effect the ASR accuracies significantly.
Artificially lowering the fundamental frequency
degrades the ASR performance marginally but this drop
in performance can be overcome to some extent using
Vocal Tract Length Normalisation (VTLN). Overall, we
observe that the changes in the voice source parameters
do not have a significant impact on ASR performance.
Comparison of the likelihood scores of all the phonemes
for the two age groups show that there is a systematic
mismatch in the acoustic space of the two age groups.
Comparison of the phoneme recognition rates show that
mid vowels, nasals and phonemes that depend on the
ability to create constrictions with tongue tip for
articulation are more affected by ageing than other
phonemes.},
doi = {10.1155/2010/525783},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2010/vipperla-eurasip10.pdf},
url = {http://dx.doi.org/10.1155/2010/525783},
year = 2010
}
@article{frankel07:ldm,
author = {Frankel, J. and King, S.},
title = {Speech Recognition using Linear Dynamic Models},
journal = {IEEE {T}ransactions on {S}peech and {A}udio
{P}rocessing},
volume = 15,
number = 1,
pages = {246--256},
abstract = {The majority of automatic speech recognition (ASR)
systems rely on hidden Markov models, in which Gaussian
mixtures model the output distributions associated with
sub-phone states. This approach, whilst successful,
models consecutive feature vectors (augmented to
include derivative information) as statistically
independent. Furthermore, spatial correlations present
in speech parameters are frequently ignored through the
use of diagonal covariance matrices. This paper
continues the work of Digalakis and others who proposed
instead a first-order linear state-space model which
has the capacity to model underlying dynamics, and
furthermore give a model of spatial correlations. This
paper examines the assumptions made in applying such a
model and shows that the addition of a hidden dynamic
state leads to increases in accuracy over otherwise
equivalent static models. We also propose a
time-asynchronous decoding strategy suited to
recognition with segment models. We describe
implementation of decoding for linear dynamic models
and present TIMIT phone recognition results.},
categories = {am,asr,ldm,timit,search,edinburgh},
month = {January},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Frankel_King_IEEE2007.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/Frankel_King_IEEE2007.ps},
year = 2007
}
@inproceedings{wang_std_covariance_icassp2010,
author = {Wang, Dong and King, Simon and Frankel, Joe and Bell,
Peter},
title = {Stochastic Pronunciation Modelling and Soft Match for
Out-of-vocabulary Spoken Term Detection},
booktitle = {Proc. ICASSP},
address = {Dallas, Texas, USA},
abstract = {A major challenge faced by a spoken term detection
(STD) system is the detection of out-of-vocabulary
(OOV) terms. Although a subword-based STD system is
able to detect OOV terms, performance reduction is
always observed compared to in-vocabulary terms. One
challenge that OOV terms bring to STD is the
pronunciation uncertainty. A commonly used approach to
address this problem is a soft matching procedure,and
the other is the stochastic pronunciation modelling
(SPM) proposed by the authors. In this paper we compare
these two approaches, and combine them using a
discriminative decision strategy. Experimental results
demonstrated that SPM and soft match are highly
complementary, and their combination gives significant
performance improvement to OOV term detection.},
keywords = {confidence estimation, spoken term detection, speech
recognition},
month = mar,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2010/wang10_icassp.pdf},
year = 2010
}
@inproceedings{livescu07:manual,
author = {Livescu, K. and Bezman, A. and Borges, N. and Yung, L.
and Çetin, Ö. and Frankel, J. and King, S. and
Magimai-Doss, M. and Chi, X. and Lavoie, L.},
title = {Manual transcription of conversational speech at the
articulatory feature level},
booktitle = {Proc. ICASSP},
address = {Honolulu},
abstract = {We present an approach for the manual labeling of
speech at the articulatory feature level, and a new set
of labeled conversational speech collected using this
approach. A detailed transcription, including
overlapping or reduced gestures, is useful for studying
the great pronunciation variability in conversational
speech. It also facilitates the testing of feature
classiers, such as those used in articulatory
approaches to automatic speech recognition. We describe
an effort to transcribe a small set of utterances drawn
from the Switchboard database using eight articulatory
tiers. Two transcribers have labeled these utterances
in a multi-pass strategy, allowing for correction of
errors. We describe the data collection methods and
analyze the data to determine how quickly and reliably
this type of transcription can be done. Finally, we
demonstrate one use of the new data set by testing a
set of multilayer perceptron feature classiers against
both the manual labels and forced alignments.},
month = {April},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/livescu_icassp07_trans.pdf},
year = 2007
}
@inproceedings{frankel00:NN_LDM,
author = {Frankel, J. and Richmond, K. and King, S. and Taylor,
P.},
title = {An automatic speech recognition system using neural
networks and linear dynamic models to recover and model
articulatory traces},
booktitle = {Proc. {ICSLP}},
abstract = {In this paper we describe a speech recognition system
using linear dynamic models and articulatory features.
Experiments are reported in which measured articulation
from the MOCHA corpus has been used, along with those
where the articulatory parameters are estimated from
the speech signal using a recurrent neural network.},
categories = {am,artic,asr,ldm,mocha,edinburgh,inversion,ann},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2000/Frankel_et_al_ICSLP2000.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2000/Frankel_et_al_ICSLP2000.ps},
year = 2000
}
@inproceedings{joe_dong_simon_interspeech08_bottle,
author = {Joe Frankel and Dong Wang and Simon King},
title = {Growing bottleneck features for tandem {ASR}},
booktitle = {Proc. Interspeech},
pages = {1549},
abstract = { We present a method for training bottleneck MLPs for
use in tandem ASR. Experiments on meetings data show
that this approach leads to improved performance
compared with training MLPs from a random
initialization. },
categories = {tandem ASR, bottleneck MLP},
month = sep,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2008/bottlenet.a.pdf},
year = 2008
}
@inproceedings{dongwang_interspeech09_cmb,
author = {Javier Tejedor and Dong Wang and Simon King and Joe
Frankel and Jose Colas},
title = {A Posterior Probability-based System Hybridisation and
Combination for Spoken Term Detection},
booktitle = {Proc. Interspeech},
pages = {2131--2134},
address = {Brighton, UK},
abstract = {Spoken term detection (STD) is a fundamental task for
multimedia information retrieval. To improve the
detection performance, we have presented a direct
posterior-based confidence measure generated from a
neural network. In this paper, we propose a
detection-independent confidence estimation based on
the direct posterior confidence measure, in which the
decision making is totally separated from the term
detection. Based on this idea, we first present a
hybrid system which conducts the term detection and
confidence estimation based on different sub-word
units, and then propose a combination method which
merges detections from heterogeneous term detectors
based on the direct posterior-based confidence.
Experimental results demonstrated that the proposed
methods improved system performance considerably for
both English and Spanish. },
categories = {joint-multigram, pronunciation model, spoken term
detection, speech recognition},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2009/cmb.pdf},
year = 2009
}
@inproceedings{toth:frankel:goztolya:king:interspeech2008,
author = {Laszlo Toth and Joe Frankel and Gabor Gosztolya and
Simon King},
title = {Cross-lingual Portability of MLP-Based Tandem Features
-- A Case Study for English and Hungarian},
booktitle = {Proc. Interspeech},
pages = {2695-2698},
address = {Brisbane, Australia},
abstract = {One promising approach for building ASR systems for
less-resourced languages is cross-lingual adaptation.
Tandem ASR is particularly well suited to such
adaptation, as it includes two cascaded modelling
steps: feature extraction using multi-layer perceptrons
(MLPs), followed by modelling using a standard HMM. The
language-specific tuning can be performed by adjusting
the HMM only, leaving the MLP untouched. Here we
examine the portability of feature extractor MLPs
between an Indo-European (English) and a Finno-Ugric
(Hungarian) language. We present experiments which use
both conventional phone-posterior and articulatory
feature (AF) detector MLPs, both trained on a much
larger quantity of (English) data than the monolingual
(Hungarian) system. We find that the cross-lingual
configurations achieve similar performance to the
monolingual system, and that, interestingly, the AF
detectors lead to slightly worse performance, despite
the expectation that they should be more
language-independent than phone-based MLPs. However,
the cross-lingual system outperforms all other
configurations when the English phone MLP is adapted on
the Hungarian data. },
keywords = {tandem, ASR},
month = {September},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2008/IS080729.PDF},
year = 2008
}
@article{tejedor:wang:frankel:king:colas:specom2008,
author = {Javier Tejedor and Dong Wang and Joe Frankel and Simon
King and José Colás},
title = {A comparison of grapheme and phoneme-based units for
{S}panish spoken term detection},
journal = {Speech Communication},
volume = {50},
number = {11-12},
pages = {980-991},
abstract = {The ever-increasing volume of audio data available
online through the world wide web means that automatic
methods for indexing and search are becoming essential.
Hidden Markov model (HMM) keyword spotting and lattice
search techniques are the two most common approaches
used by such systems. In keyword spotting, models or
templates are defined for each search term prior to
accessing the speech and used to find matches. Lattice
search (referred to as spoken term detection), uses a
pre-indexing of speech data in terms of word or
sub-word units, which can then quickly be searched for
arbitrary terms without referring to the original
audio. In both cases, the search term can be modelled
in terms of sub-word units, typically phonemes. For
in-vocabulary words (i.e. words that appear in the
pronunciation dictionary), the letter-to-sound
conversion systems are accepted to work well. However,
for out-of-vocabulary (OOV) search terms,
letter-to-sound conversion must be used to generate a
pronunciation for the search term. This is usually a
hard decision (i.e. not probabilistic and with no
possibility of backtracking), and errors introduced at
this step are difficult to recover from. We therefore
propose the direct use of graphemes (i.e., letter-based
sub-word units) for acoustic modelling. This is
expected to work particularly well in languages such as
Spanish, where despite the letter-to-sound mapping
being very regular, the correspondence is not
one-to-one, and there will be benefits from avoiding
hard decisions at early stages of processing. In this
article, we compare three approaches for Spanish
keyword spotting or spoken term detection, and within
each of these we compare acoustic modelling based on
phone and grapheme units. Experiments were performed
using the Spanish geographical-domain Albayzin corpus.
Results achieved in the two approaches proposed for
spoken term detection show us that trigrapheme units
for acoustic modelling match or exceed the performance
of phone-based acoustic models. In the method proposed
for keyword spotting, the results achieved with each
acoustic model are very similar.},
categories = {Spoken term detection; Keyword spotting; Graphemes;
Spanish},
doi = {10.1016/j.specom.2008.03.005},
month = {November-December},
year = 2008
}
@inproceedings{king00:recognition_syll,
author = {King, S. and Taylor, P. and Frankel, J. and Richmond,
K.},
title = {Speech recognition via phonetically-featured syllables},
booktitle = {PHONUS},
volume = {5},
pages = {15-34},
address = {Institute of Phonetics, University of the Saarland},
abstract = {We describe recent work on two new automatic speech
recognition systems. The first part of this paper
describes the components of a system based on
phonological features (which we call EspressoA) in
which the values of these features are estimated from
the speech signal before being used as the basis for
recognition. In the second part of the paper, another
system (which we call EspressoB) is described in which
articulatory parameters are used instead of
phonological features and a linear dynamical system
model is used to perform recognition from automatically
estimated values of these articulatory parameters.},
categories = {am,artic,asr,ldm,phonetic_feature,mocha,timit,edinburgh},
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2000/King_et_al_Phonus2000.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2000/King_et_al_Phonus2000.ps},
year = 2000
}
@inproceedings{frankel05:hybrid,
author = {Frankel, J. and King, S.},
title = {A Hybrid {ANN/DBN} Approach to Articulatory Feature
Recognition},
booktitle = {Proc. Eurospeech},
address = {Lisbon},
abstract = {Artificial neural networks (ANN) have proven to be
well suited to the task of articulatory feature (AF)
recognition. Previous studies have taken a cascaded
approach where separate ANNs are trained for each
feature group, making the assumption that features are
statistically independent. We address this by using
ANNs to provide virtual evidence to a dynamic Bayesian
network (DBN). This gives a hybrid ANN/DBN model and
allows modelling of inter-feature dependencies. We
demonstrate significant increases in AF recognition
accuracy from modelling dependencies between features,
and present the results of embedded training
experiments in which a set of asynchronous feature
changes are learned. Furthermore, we report on the
application of a Viterbi training scheme in which we
alternate between realigning the AF training labels and
retraining the ANNs.},
categories = {am,artic,asr,dbn,oginumbers,edinburgh},
month = sep,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2005/Frankel_King_INTER2005.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2005/Frankel_King_INTER2005.ps},
year = 2005
}
@phdthesis{frankel03:thesis,
author = {Frankel, J.},
title = {Linear dynamic models for automatic speech recognition},
school = {The Centre for Speech Technology Research, Edinburgh
University},
abstract = {The majority of automatic speech recognition (ASR)
systems rely on hidden Markov models (HMM), in which
the output distribution associated with each state is
modelled by a mixture of diagonal covariance Gaussians.
Dynamic information is typically included by appending
time-derivatives to feature vectors. This approach,
whilst successful, makes the false assumption of
framewise independence of the augmented feature vectors
and ignores the spatial correlations in the
parametrised speech signal. This dissertation seeks to
address these shortcomings by exploring acoustic
modelling for ASR with an application of a form of
state-space model, the linear dynamic model (LDM).
Rather than modelling individual frames of data, LDMs
characterize entire segments of speech. An
auto-regressive state evolution through a continuous
space gives a Markovian model of the underlying
dynamics, and spatial correlations between feature
dimensions are absorbed into the structure of the
observation process. LDMs have been applied to speech
recognition before, however a smoothed Gauss-Markov
form was used which ignored the potential for subspace
modelling. The continuous dynamical state means that
information is passed along the length of each segment.
Furthermore, if the state is allowed to be continuous
across segment boundaries, long range dependencies are
built into the system and the assumption of
independence of successive segments is loosened. The
state provides an explicit model of temporal
correlation which sets this approach apart from
frame-based and some segment-based models where the
ordering of the data is unimportant. The benefits of
such a model are examined both within and between
segments. LDMs are well suited to modelling smoothly
varying, continuous, yet noisy trajectories such as
found in measured articulatory data. Using
speaker-dependent data from the MOCHA corpus, the
performance of systems which model acoustic,
articulatory, and combined acoustic-articulatory
features are compared. As well as measured articulatory
parameters, experiments use the output of neural
networks trained to perform an articulatory inversion
mapping. The speaker-independent TIMIT corpus provides
the basis for larger scale acoustic-only experiments.
Classification tasks provide an ideal means to compare
modelling choices without the confounding influence of
recognition search errors, and are used to explore
issues such as choice of state dimension, front-end
acoustic parametrization and parameter initialization.
Recognition for segment models is typically more
computationally expensive than for frame-based models.
Unlike frame-level models, it is not always possible to
share likelihood calculations for observation sequences
which occur within hypothesized segments that have
different start and end times. Furthermore, the Viterbi
criterion is not necessarily applicable at the frame
level. This work introduces a novel approach to
decoding for segment models in the form of a stack
decoder with $A^*$ search. Such a scheme allows
flexibility in the choice of acoustic and language
models since the Viterbi criterion is not integral to
the search, and hypothesis generation is independent of
the particular language model. Furthermore, the
time-asynchronous ordering of the search means that
only likely paths are extended, and so a minimum number
of models are evaluated. The decoder is used to give
full recognition results for feature-sets derived from
the MOCHA and TIMIT corpora. Conventional train/test
divisions and choice of language model are used so that
results can be directly compared to those in other
studies. The decoder is also used to implement Viterbi
training, in which model parameters are alternately
updated and then used to re-align the training data.},
categories = {am,artic,asr,ldm,mocha,timit,search,edinburgh},
month = apr,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2003/Frankel_thesis2003.pdf},
ps = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2003/Frankel_thesis2003.ps},
year = 2003
}
@article{king07:JASA2007,
author = {King, S. and Frankel, J. and Livescu, K. and
McDermott, E. and Richmond, K. and Wester, M.},
title = {Speech production knowledge in automatic speech
recognition},
journal = {Journal of the Acoustical Society of America},
volume = 121,
number = 2,
pages = {723--742},
abstract = {Although much is known about how speech is produced,
and research into speech production has resulted in
measured articulatory data, feature systems of
different kinds and numerous models, speech production
knowledge is almost totally ignored in current
mainstream approaches to automatic speech recognition.
Representations of speech production allow simple
explanations for many phenomena observed in speech
which cannot be easily analyzed from either acoustic
signal or phonetic transcription alone. In this
article, we provide a survey of a growing body of work
in which such representations are used to improve
automatic speech recognition.},
month = feb,
pdf = {http://www.cstr.inf.ed.ac.uk/downloads/publications/2007/King_et_al_review.pdf},
year = 2007
}
![[ intranet ]](/images/menu_intranet.png)
![[ festival ]](/images/menu_festival.png)