# sjg.bib

@INPROCEEDINGS{Rayner_Godsill_1991,
AUTHOR = {P. J. W.  Rayner and
S. J.  Godsill},
TITLE = {The Detection and Correction of Artefacts
in Archived Gramophone Recordings},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
MONTH = OCT,
YEAR = 1991,
ABSTRACT = { This
paper presents recent developments in techniques for the restoration
associated with all forms of audio media, including CD and DAT, but
is most characteristic of gramophone disks. One limitation of
restoration performance is observed when the length of a click
becomes large. Visual examination of waveforms shows that the
restored signal often does not have enough energy towards the centre
of the gap. For many audio signals the effect is visible for scratch
lengths greater than 30 samples. Fortunately, the problem is
generally not audible until much longer scratches are interpolated,
greater than say 100 samples. This phenomenon is a major limiting
factor on the maximum number of samples which the process may
successfully interpolate... }
}


@INPROCEEDINGS{Godsill_Rayner_1992,
AUTHOR = {S. J.  Godsill and P. J. W.  Rayner},
TITLE = {A {B}ayesian Approach to the Detection and Correction of
Bursts of Errors in Audio Signals},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = 2,
PAGES = {261-264},
MONTH = MAR,
YEAR = 1992,
ABSTRACT = {
In this paper we derive the {\em a posteriori} probability for the
location of bursts of noise additively superimposed on a Gaussian AR
process. The MAP solution for noise burst position is obtained by
using a simple search procedure, yielding the noise burst location
corresponding to minimum probability of error. This procedure finds
application in digital audio processing, where clicks and scratches
may be modelled as additive bursts of noise. The method enables
accurate detection of these degradations and their subsequent
replacement (interpolation). Experiments are carried out on both
real audio data and synthetic AR processes, and comparisons are made
with existing techniques.
}
}


@INPROCEEDINGS{Godsill_Rayner_1993a,
AUTHOR = {S. J.  Godsill and P. J. W.  Rayner},
TITLE = {Frequency-Domain Interpolation of Sampled Signals},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = {I},
PAGES = {209-212},
YEAR = 1993,
MONTH = APR,
ABSTRACT = { A new interpolation technique has been developed for replacing
missing samples in a sampled waveform drawn from a stationary
stochastic process, given the power spectrum for the process. The
method works with a finite block of data and is based on the
assumption that components of the block DFT are Gaussian zero-mean
independent random variables with variance proportional to the power
spectrum at each frequency value. These assumptions make the
interpolator particularly suitable for signals with a
sharply-defined harmonic structure, such as audio waveforms recorded
from music or voiced speech. Some results are presented and
comparisons are made with existing techniques.}
}


@INPROCEEDINGS{Godsill_Rayner_1993b,
AUTHOR = {S. J.  Godsill and P. J. W. Rayner},
TITLE = {The Restoration of Pitch Variation Defects in Gramophone Recordings},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 1993,
MONTH = OCT,
ABSTRACT = {
A new algorithm is presented for the identification and restoration
of time-varying pitch defects in audio signals. The problem is
commonly encountered as wow' in gramophone disc and magnetic tape
recordings where motor speed variations or eccentricity in the
recording process are significant. The algorithm operates in two
stages, the first of which tracks tonal components in musical
signals to generate a single {\em pitch variation curve}, and the
second stage which performs restoration as a time-varying resampling
operation. Results are presented from both artificially degraded
sources and real sources.
}
}


@INPROCEEDINGS{Godsill_1994a,
AUTHOR = {S. J. Godsill},
TITLE = {Recursive Restoration of Pitch Variation Defects in Musical Recordings},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = 2,
PAGES = {233-236},
YEAR = 1994,
MONTH = {April},
ABSTRACT = {
A new algorithm is presented for the identification and restoration
of time-varying pitch defects in audio signals. The problem is
commonly encountered as wow' in gramophone disc and magnetic tape
recordings where motor speed variations or eccentricity in the
recording process are significant. The algorithm operates in two
stages, the first of which tracks tonal components in musical
signals to generate a single pitch variation curve, and the second
stage which performs restoration as a time-varying resampling
operation. Results are presented from both artificially degraded
sources and real sources. },
KEYWORDS = {wow, flutter, smoothness, Bayesian regularization}
}


@INPROCEEDINGS{Hicks_Godsill_1994,
AUTHOR = {C. M.  Hicks and S. J.  Godsill},
TITLE = {A 2-Channel Approach to the Removal of Impulsive Noise from archived recordings},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = 2,
PAGES = {213--216},
YEAR = 1994,
MONTH = APR,
ABSTRACT = {
In this paper we discuss the extraction of two signals from a
monophonic gramophone record, and observe that the impulsive noise
is significantly different in the two.  This, coupled with the
redundancy of the desired parts of the signals, has great advantages
in the processes of impulsive noise detection and removal.  It is
simple to obtain two suitable signals by using a stereo replay
cartridge, and we develop detection and interpolation algorithms
that use such a pair of signals.  We present the results of computer
simulations and informal listening tests on archive material.  In
both cases the two-channel method is found to be an improvement over
a similar single-channel algorithm, with little change in
computational cost.
}
}


@INPROCEEDINGS{Godsill_Rayner_1995d,
AUTHOR = {S. J.  Godsill and P. J. W.  Rayner},
TITLE = {Robust Noise Modelling with Application to Audio Restoration},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 1995,
MONTH = OCT,
ABSTRACT = {
New methods are presented for robust modelling of noise sources
drawn from heavy-tailed or impulsive distributions, such as are
commonly encountered in communications systems and corrupted audio
signals. The methods are formulated for linear signal models within
a Bayesian framework (although likelihood-based results are easily
obtained as a subset of the Bayesian methods). Solutions are
generated using powerful iterative techniques. Investigations are
carried out for an audio interpolation framework in which certain
samples are corrupted with additive impulsive noise.
}
}


@INPROCEEDINGS{Chan_Rayner_Godsill_1995,
AUTHOR = {D. C. B. Chan and P. J. W. Rayner and S. J. Godsill},
TITLE = {Multi-channel Blind Signal Separation By Decorrelation},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 1995,
MONTH = OCT,
PDF = {http://www2.eng.cam.ac.uk/~dcbc1/research/mohonk95.zip},
ABSTRACT = {
The separation of independent sources from mixed observed data is a
fundamental and challenging problem. In many practical situations,
observations may be modelled as linear mixtures of a number of
source signals, ie. a linear multi-input multi-output system. A
typical example is speech recordings made in an acoustic environment
in the presence of background noise and/or competing speakers. Other
examples include EEG signals, passive sonar applications and
cross-talk in data communications. In this paper, we propose
iterative algorithms to solve the $n \times n$ linear time invariant
system under two different constraints. Some existing solutions for
$2 \times 2$ systems are reviewed and compared.
}
}


@INPROCEEDINGS{Godsill_Rayner_1996a,
AUTHOR = {S. J. Godsill and P. J. W. Rayner},
TITLE = { Robust noise reduction for speech and audio signals },
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 1996,
MONTH = MAY,
ABSTRACT = {
Statistical model-based methods are presented for the reconstruction
of autocorrelated signals in impulsive plus continuous noise
environments. Signals are modelled as autoregressive and noise
sources as discrete and continuous mixtures of Gaussians, allowing
for robustness in highly impulsive and non-Gaussian environments.
Markov Chain Monte Carlo methods are used for reconstruction of the
corrupted waveforms within a Bayesian probabilistic framework and
results are presented for contaminated voice and audio signals.
}
}


@INPROCEEDINGS{Chan_Rayner_Godsill_1996a,
AUTHOR = {D. C. B. Chan and P. J. W. Rayner and S. J. Godsill},
TITLE = { Multi-channel blind signal separation },
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
PDF = {http://www2.eng.cam.ac.uk/~dcbc1/research/icassp96.zip},
YEAR = 1996,
MONTH = MAY
}


@INPROCEEDINGS{Godsill_Kokaram_1996a,
AUTHOR = {S. J. Godsill and A. C. Kokaram},
TITLE = { Joint Interpolation, motion and parameter estimation for
image
sequences with missing data.
},
BOOKTITLE = {Proc. EUSIPCO},
YEAR = 1996,
MONTH = SEP,
PDF = {http://www-sigproc.eng.cam.ac.uk/~ack/acksjg.ps.gz}
}


@INPROCEEDINGS{Troughton_Godsill_1997a,
AUTHOR = {P. T. Troughton and S. J. Godsill},
TITLE = { Bayesian Model Selection for time series using {M}arkov
Chain {M}onte {C}arlo},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 1997,
MONTH = APR,
PDF = {http://www-sigproc.eng.cam.ac.uk/\~ptt10/papers/icassp97.ps.gz}
}


@INPROCEEDINGS{Godsill_1997a,
AUTHOR = {S. J. Godsill},
TITLE = { Robust modelling of noisy {ARMA} signals },
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 1997,
MONTH = APR
}


@INPROCEEDINGS{Campbell_Godsill_1998,
TITLE = {On a new stochastic version of the {EM} algorithm},
AUTHOR = {C. Campbell and S. J. Godsill},
BOOKTITLE = {Proc. European
Conference on Signal Processing},
YEAR = 1998,
MONTH = SEP
}


@INPROCEEDINGS{Troughton_Godsill_1998b,
AUTHOR = {P.T. Troughton and
S. J. Godsill},
TITLE = {A Reversible Jump Sampler for Autoregressive
Time Series},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = {1998},
VOLUME = {IV},
PAGES = {2257-2260},
MONTH = APR,
NOTES = {ISBN 0 876346 10
8},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~ptt10/papers/icassp98.pdf}
}


@INPROCEEDINGS{Troughton_Godsill_1998a,
TITLE = {{MCMC} methods for restoration of nonlinearly distorted autoregressive
signals},
AUTHOR = {P.T. Troughton and  S. J. Godsill},
BOOKTITLE = {Proc. European
Conference on Signal Processing},
YEAR = 1998,
MONTH = SEP
}


@INPROCEEDINGS{Walmsley_Godsill_Rayner_1998,
TITLE = {Multidimensional optimisation of harmonic signals},
AUTHOR = {P.J. Walmsley and S. J. Godsill and P. J. W.  Rayner},
BOOKTITLE = {Proc. European
Conference on Signal Processing},
YEAR = 1998,
MONTH = SEP
}


@INPROCEEDINGS{Godsill_Tan_1997,
AUTHOR = {S. J.  Godsill and C. H. Tan},
TITLE = {Removal of low frequency transient noise from old recordings using
model-based signal separation techniques},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 1997,
MONTH = OCT
}


@INPROCEEDINGS{Kokaram_Godsill_1998,
TITLE = {Joint noise reduction, motion estimation, missing data reconstruction, and model parameter estimation for degraded motion pictures},
AUTHOR = {A.C. Kokaram and  S. J. Godsill},
BOOKTITLE = {Proc. SPIE, San Diego},
YEAR = 1998,
MONTH = JUL
}


@INPROCEEDINGS{Andrieu_Doucet_Godsill_1998,
TITLE = {Bayesian blind marginal separation of convolutively mixed discrete sources},
AUTHOR = {C. Andrieu and A. Doucet and  S. J. Godsill},
BOOKTITLE = {Proc. IEEE Workshop - Neural networks for signal processing, Cambridge},
YEAR = 1998,
MONTH = AUG
}


@INPROCEEDINGS{Godsill_1999a,
AUTHOR = {S. J. Godsill},
TITLE = {{MCMC}
and {EM}-based methods for inference in heavy-tailed processes with
alpha-stable innovations},
BOOKTITLE = {Proc. IEEE Signal processing
workshop on higher-order statistics},
YEAR = 1999,
ABSTRACT = {In this
paper we present both stochastic and deterministic iterative methods
for inference about random processes with symmetric stable
innovations. The proposed methods use a scale mixtures of normals
(SMiN) representation of the symmetric stable law to express the
processes in conditionally Gaussian form. This allows standard
procedures for dealing with the Gaussian case to be re-used directly
as part of the scheme.  In contrast with other recently published
work on the topic, we propose a novel hybrid rejection sampling
method for simulating the scale parameters from their full
conditional distributions, making use of asymptotic approximations
for the tail of a positive stable distribution when rejection rates
are too high. This hybrid approach potentially leads to improved
performance compared with straightforward rejection sampling or
Metropolis-Hastings (M-H) approaches. The methods can be applied to
any model with symmetric stable terms, but we illustrate their
application to linear models and present simulations for AR time
series with stable innovations.},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~{}sjg/papers/99/stable.ps},
MONTH = JUN,
NOTE = {Caesarea, Israel}
}


@INPROCEEDINGS{Godsill_Kuruoglu_1999,
AUTHOR = {S. J. Godsill and E.
E. Kuruoglu},
TITLE = {Bayesian inference for time series with
heavy-tailed
symmetric $\alpha$-stable noise processes},
BOOKTITLE = {Proc. Applications of heavy tailed distributions in
economics, engineering and statistics},
ABSTRACT = {In this paper we
present both stochastic and deterministic iterative methods for
inference about random processes with symmetric stable innovations.
The proposed methods use a scale mixtures of normals (SMiN)
representation of the symmetric stable law to express the processes
in conditionally Gaussian form. This allows standard procedures for
dealing with the Gaussian case to be re-used directly as part of the
scheme. In contrast with other recently published work on the topic,
we propose a novel hybrid rejection sampling method for simulating
the scale parameters from their full conditional distributions,
making use of asymptotic approximations for the tail of a positive
stable distribution when rejection rates are too high. This hybrid
approach leads to improved performance compared with straightforward
rejection sampling or Metropolis-Hastings (M-H) approaches. The
methods can be applied to any model with symmetric stable terms, but
we illustrate their application to linear models and present
simulations for AR time series with symmetric stable innovations},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/99/symm_stab.ps},
YEAR = 1999,
MONTH = JUN,
NOTE = {Washington DC, USA}
}


@INPROCEEDINGS{Djuric_Godsill_Fitzgerald_Rayner_1998,
AUTHOR = {P.M.
Djuri\'{c} and S. J. Godsill and  W. J. Fitzgerald and P. J. W.
Rayner},
TITLE = {Detection and estimation of signals by reversible
jump {M}arkov chain {M}onte {C}arlo computations},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = 4,
PAGES = {2269--2272},
YEAR = 1998,
NOTE = {Seattle}
}


@ARTICLE{troughton99nsip,
AUTHOR = {Paul T. Troughton and Simon J. Godsill},
TITLE = {{MCMC} Methods for Restoration of Quantised Time Series},
YEAR = 1999,
JOURNAL = {Proceedings of the
IEEE-EURASIP Workshop on Nonlinear Signal and Image Processing},
ABSTRACT = {
In digital systems, the amplitude of a time series is quantised with
finite resolution. This is a nonlinear process which introduces
distortion.

We develop a Bayesian, model-based approach to reducing the
quantisation distortion when moving a time series, such as an audio
signal, to a higher resolution medium. The signal is modelled as a
discrete-time, continuous-valued autoregressive (AR) process of
unknown order.

The model parameters and reconstructed signal are estimated using
Markov chain Monte Carlo (MCMC) techniques. This requires samples to
be drawn from a truncated multivariate Gaussian distribution, for
which a Metropolis-Hastings approach is developed. },
VOLUME = 2,
PAGES = {447--451},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/99/nsip99.ps}
}


@INPROCEEDINGS{Clapp_Godsill_1999b,
AUTHOR = {T. C. Clapp and S. J.
Godsill},
TITLE = {Fixed-lag Blind Equalization and Sequence
Estimation in Digital Communications Systems using Sequential
Importance Sampling},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
NOTE = {Arizona},
MONTH = MAR,
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/99/tcc_icassp99_ps.gz},
PAGES = {2495--2498},
VOLUME = 5,
YEAR = 1999
}


@INPROCEEDINGS{Djuric_Godsill_1998,
AUTHOR = {P.M. Djuric and S.J.
Godsill},
YEAR = 1998,
TITLE = {Parametric modeling and estimation of
time varying spectra},
BOOKTITLE = {Proc. Asilomar Conference on
Signals and Systems}
}


@INPROCEEDINGS{Robert_Doucet_Godsill_1999,
AUTHOR = {C. P.  Robert and A. Doucet and S. J. Godsill},
TITLE = {Maximization of Marginal Posterior Distributions using
{M}arkov Chain {M}onte {C}arlo Methods},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
NOTE = {Arizona},
MONTH = MAR,
ABSTRACT = {Markov chain Monte Carlo (MCMC)
methods are powerful \linebreak simulation-based techniques for
sampling from high-dimensional and/or non-standard probability
distributions. These methods have recently become very popular in
the statistical and signal processing communities as they allow
highly complex inference problems in detection and estimation to be
problem of marginal \textit{maximum a posteriori} (MMAP) estimation.
In this paper, we present a simple and novel MCMC strategy, called
State-Augmentation for Marginal Estimation (SAME), that allows MMAP
estimates to be obtained for Bayesian models. The methodology is
very general and we illustrate the simplicity and utility of the
approach by examples in MAP\ parameter estimation for Hidden Markov
models (HMMs) and for missing data interpolation in autoregressive
time series.},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/98/mcmc_opt.ps.gz},
VOLUME = 3,
PAGES = {1753--1756},
YEAR = 1999
}


@INPROCEEDINGS{Andrieu_Godsill_1999,
AUTHOR = {C. Andrieu and S. J.
Godsill},
TITLE = {Bayesian Separation and Recovery of Convolutively
Mixed Autoregressive Sources},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/99/sigsep.ps},
NOTE = {Arizona},
MONTH = MAR,
VOLUME = 3,
PAGES = {1733--1736},
ABSTRACT = {In this paper we address the problem of the separation and
recovery of convolutively mixed autoregressive processes in a
Bayesian framework. Solving this problem requires the ability to solve integration and/or optimization problems of
complicated posterior distributions. We thus propose efficient stochastic algorithms based on Markov chain Monte Carlo                               (MCMC) methods. We present three algorithms. The first one is a classical Gibbs sampler that generates samples from
the posterior distribution. The two other algorithms are stochastic optimization algorithms that allow to optimize either the
marginal distribution of the source s, or the marginal distribution of the parameters of the sources and mixing filters,
conditional upon the observation. Simulations are presented.},
YEAR = 1999
}


@INPROCEEDINGS{Ahmed_Rayner_godsill_1999a,
AUTHOR = {A Ahmed and  P. J.W. Rayner and  S. J. Godsill},
TITLE = {CONSIDERING NON-STATIONARITY FOR BLIND SIGNAL SEPARATION},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
ABSTRACT = {We investigate the exploitation of non-stationarity for signal separation. A
second-order decorrelation method is used to separate synthetic
independent autoregressive signals that are made up of stationary
blocks that have been convolutively mixed. We compare results
obtained by not taking into account the non-stationarity with those
that do. Under certain conditions, exploiting non- stationarity
results in more robust separation. We present simulation results
that vindicate this fact. In addition, we apply the decorrelation
method to real microphone signals, to see how exploiting
non-stationarity affects separation quality.},
YEAR = 1999,
MONTH = OCT
}


@INPROCEEDINGS{Ahmed_Rayner_godsill_1999b,
AUTHOR = {A Ahmed and  P. J.W. Rayner and  S. J. Godsill},
TITLE = {Recursive Decorrelation For Blind Convolutive signal Separation},
BOOKTITLE = {Proc. Independent component analysis and blind signal separation, Aussois, France},
ABSTRACT = {We present a second-order statistics based method for the blind
separation of convolutively mixed audio signals. A cost function,
which is the sum of the squares of the cross-correlations of the
source estimates for an arbitrary number of lags, is minimised using
Newton's method. The method relies on the assumption that the
sources have non-zero temporal correlations. Results are presented
for synthetically mixed real sources},
YEAR = 1999,
MONTH = JAN
}


@INPROCEEDINGS{Walmsley_Godsill_Rayner_1999,
AUTHOR = {P. J. Walmsley and S. J. Godsill and P. J. W.
Rayner},
TITLE = { POLYPHONIC PITCH TRACKING USING JOINT {B}AYESIAN ESTIMATION OF MULTIPLE FRAME PARAMETERS},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 1999,
MONTH = OCT
}


@INPROCEEDINGS{Godsill_Doucet_West_00a,
AUTHOR = {S J Godsill and A Doucet and M West},
BOOKTITLE = {Proc. International Symposium on Frontiers of Time Series Modelling},
TITLE = {Methodology for {M}onte {C}arlo smoothing with application to time-varying autoregressions},
MONTH = FEB,
NOTE = {Institute of Statistical Mathematics, Tokyo},
YEAR = {2000},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/00/smoother.ps}
}


@INPROCEEDINGS{Godsill_Doucet_West_00b,
AUTHOR = {A. Doucet and S. J. Godsill and M. West},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
TITLE = {{M}ONTE {C}ARLO FILTERING AND SMOOTHING WITH APPLICATION TO TIME-VARYING SPECTRAL ESTIMATION},
YEAR = {2000},
VOLUME = {II},
PAGES = {701--704},
NOTE = {ISBN 0-7803-6296-9},
ABSTRACT = {We develop methods for performing filtering and smoothing
in non-linear non-Gaussian dynamical models. The methods rely on a
particle cloud representation of the filtering distribution which
evolves through time using importance sampling and resampling ideas.
In particular, novel techniques are presented for generation of
random realisations from the joint smoothing distribution and for
MAP estimation of the state sequence. Realisations of the smoothing
distribution are generated in a forward-backward procedure, while
the MAP estimation procedure can be performed in a single forward
pass of the Viterbi algorithm applied to a discretised version of
the state space. An application to spectral estimation for
time-varying autoregressions is described.},
PDF = {http://www-sigproc.eng.cam.ac.uk/\~sjg/papers/00/smoother1.ps}
}


@INPROCEEDINGS{Wolfe00b,
AUTHOR = {P. J. Wolfe and S. J. Godsill},
TITLE = {Towards a Perceptually Optimal Spectral Amplitude
Estimator for Audio Signal Enhancement},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
NOTE = {ISBN 0-7803-6296-9},
VOLUME = {II},
PAGES = {821--824},
YEAR = 2000,
ABSTRACT = { We present a statistical model-based approach to
noise. Because broadband noise is localised in
neither time nor frequency, its removal is one of
the most pervasive and difficult signal enhancement
tasks. In order to improve perceived signal quality,
we take advantage of human perception and define a
best estimate of the original signal in terms of a
cost function incorporating perceptual optimality
criteria. We derive the resultant signal estimator
and implement it in a short-time spectral
attenuation framework. Audio examples, references,
and further information may be found at
http://www-sigproc.eng.cam.ac.uk/~pjw47. },
PDF = {{http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/icassp_wolfe_00.ps}
}
}


@INPROCEEDINGS{haan00:_model_elect_data_dna_sequen_mcmc,
AUTHOR = {N. M. Haan and S. J. Godsill},
TITLE = {Modelling Electropherogram Data for {DNA} Sequencing using variable dimension {MCMC}},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = {VI},
PAGES = {3542--3545},
YEAR = 2000,
NOTE = {ISBN 0-7803-6296-9},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/electro.ps}
}


@INPROCEEDINGS{Godsill_2000a,
AUTHOR = {S. J. Godsill},
TITLE = {Inference in symmetric alpha-stable noise using {MCMC}\ and the slice sampler},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
VOLUME = {VI},
PAGES = {3806--3809},
YEAR = 2000,
NOTE = {ISBN 0-7803-6296-9},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/slice.ps}
}


@INPROCEEDINGS{Fong_Godsill_2001,
AUTHOR = {W. N. W. Fong and S. J. Godsill},
TITLE = {Monte Carlo Smoothing For Non-linearly Distorted Signals},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2001,
VOLUME = 6,
PAGES = {3997--4000},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/icassp_smoother.pdf}
}


@INPROCEEDINGS{Wolfe_Dörfler_Godsill_2001,
AUTHOR = {P. J. Wolfe and
M. Dörfler and S. J. Godsill},
TITLE = {Multi-{G}abor dictionaries for
audio time-frequency analysis},
BOOKTITLE = {Proc. IEEE Workshop on Audio and
Acoustics, Mohonk, NY State},
YEAR = 2001,
MONTH = OCT,
PAGES = {43--46},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/wolfe_waspaa_01.ps}
}


@INPROCEEDINGS{Wolfe_Godsill_2001,
AUTHOR = {P.J. Wolfe and S. J.
Godsill},
TITLE = {Simple alternatives to the {E}phraim and {M}alah
suppression rule for speech enhancement},
BOOKTITLE = {Proc. IEEE Workshop on
Statistical Signal Processing},
PAGES = {496--499},
MONTH = AUG,
YEAR = 2001,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/wolfe_ssp_01.ps}
}


@INPROCEEDINGS{Haan_Godsill_2001b,
AUTHOR = {N. M. Haan and S. J.
Godsill},
TITLE = {A time-varying model for {DNA} Sequencing data
submerged in correlated noise},
BOOKTITLE = {Proc. IEEE Workshop on
Statistical Signal Processing},
MONTH = AUG,
YEAR = 2001,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/haan_ssp.pdf}
}


@INPROCEEDINGS{haan_godsill_2001a,
AUTHOR = {N.M. Haan and S.J.
Godsill},
TITLE = {SEQUENTIAL METHODS FOR {DNA} SEQUENCING},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/haan_icassp_01.pdf},
YEAR = 2001
}


@INPROCEEDINGS{Fong_Godsill_2001b,
AUTHOR = {W. Fong and S.
J. Godsill},
TITLE = {Monte Carlo smoothing with application to audio
signal enhancement},
BOOKTITLE = {Proc. IEEE Workshop on
Statistical Signal Processing},
PAGES = {18--21},
YEAR = 2001,
PDF = {http://www-sigproc.eng.cam.ac.uk/~wnwf2/Wssp01/PAPER.ps}
}


@INPROCEEDINGS{giannopoulos_godsill_2001,
AUTHOR = {P. Giannopoulos
and S. J. Godsill},
TITLE = {ESTIMATION OF {CAR} PROCESSES OBSERVED IN
NOISE USING {B}AYESIAN INFERENCE},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2001,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/giann_icassp_01.pdf}
}


@INPROCEEDINGS{Davy_Godsill_2002,
AUTHOR = {M. Davy and S. J. Godsill},
TITLE = {Detection of abrupt spectral changes using support vector machines. An application to
audio signal segmentation},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2002,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/davy_godsill_icassp02.pdf}
}


@INPROCEEDINGS{Fong_Godsill_2002a,
AUTHOR = {W. Fong and S. J. Godsill},
TITLE = {Sequential {M}onte {C}arlo simulation of dynamical models with slowly varying parameters:
application to audio},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2002,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/fong_godsill_icassp02.pdf}
}


@INPROCEEDINGS{Godsill_Davy_2002a,
AUTHOR = {S. J. Godsill and M. Davy},
TITLE = {Bayesian harmonic models for musical pitch estimation and analysis},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2002,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/godsill_davy_icassp02.pdf}
}


@INPROCEEDINGS{Haan_Godsill_2002,
AUTHOR = {N. M. Haan and S. J. Godsill},
TITLE = {Bayesian models for {DNA} sequencing},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2002,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/haan_godsill_icassp02.pdf}
}


@INPROCEEDINGS{Fong_Godsill_2002_b,
AUTHOR = {W. Fong and S. J.
Godsill},
TITLE = {Sequential {M}onte {C}arlo Simulation of Dynamical
Models with
Slowly Varying Parameters: An Extension},
BOOKTITLE = {XI European Signal Processing Conference (EUSIPCO)},
YEAR = 2002,
PDF = {http://www-sigproc.eng.cam.ac.uk/~wnwf2/Eusipco2002.pdf}
}


@INPROCEEDINGS{Wolfe_Godsill_2002a,
AUTHOR = {P. J. Wolfe and S. J.
Godsill},
TITLE = {Bayesian modelling of time-frequency coefficients
for audio signal enhancement},
EDITOR = {S. Becker and  S. Thrun and
K. Obermayer},
BOOKTITLE = {Advances in Neural Information Processing
Systems 15, Cambridge, MA},
PUBLISHER = {MIT Press},
PDF = {http://www-sigproc.eng.cam.ac.uk/~pjw47/papers/wolfe_nips_02_submit.pdf},
YEAR = 2002
}


@INPROCEEDINGS{Vermaak_Godsill_Doucet_2003a,
AUTHOR = {J.
Vermaak and S. J. Godsill and A. Doucet},
FUNCTION REGRESSION USING TRANS-DIMENSIONAL SEQUENTIAL {M}ONTE
{C}ARLO},
BOOKTITLE = {Proc. IEEE Workshop on Statistical Signal
Processing},
YEAR = 2003
}


@INPROCEEDINGS{Vermaak_Godsill_Doucet_2003b,
AUTHOR = {J. Vermaak and S. J. Godsill and A. Doucet},
TITLE = {Sequential {B}ayesian Kernel Regression},
in Neural Information Processing Systems 16, Cambridge, MA},
PUBLISHER = {MIT Press},
YEAR = 2003
}


@INPROCEEDINGS{Godsill_Vermaak_2004,
AUTHOR = {S. J. Godsill and J.
Vermaak},
TITLE = {MODELS AND ALGORITHMS FOR TRACKING USING
TRANS-DIMENSIONAL SEQUENTIAL {M}ONTE {C}ARLO},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2004
}


@INPROCEEDINGS{Kashino_Godsill_2004,
AUTHOR = {K. Kashino and S. Godsill},
TITLE = {BAYESIAN ESTIMATION OF
SIMULTANEOUS MUSICAL NOTES BASED ON FREQUENCY DOMAIN MODELLING},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2004
}


@INPROCEEDINGS{Vermaak_Ikoma_Godsill_2004,
AUTHOR = {J. Vermaak, N.
Ikoma and S. J. Godsill},
TITLE = {Extended Object Tracking using
Particle techniques},
BOOKTITLE = {Proc. IEEE Aerospace},
YEAR = 2004
}


@INPROCEEDINGS{Ikoma_Godsill_2003,
AUTHOR = {N. Ikoma and  S. J.
Godsill},
TITLE = {Extended object tracking with unknown association,
missing observations and clutter using particle filters},
BOOKTITLE = {Proc. IEEE Workshop on Statistical Signal Processing},
YEAR = 2003
}


@INPROCEEDINGS{Fevotte_Godsill_Wolfe_2004,
AUTHOR = {C\'{e}dric F\'{e}votte and Simon J. Godsill and Patrick J.
Wolfe},
TITLE = { Bayesian approach for blind separation of
underdetermined mixtures of sparse sources},
BOOKTITLE = {Proc.
Internl. Workshop on ICA},
PDF = {http://www-sigproc.eng.cam.ac.uk/~cf269/Proceedings/ica04.pdf},
YEAR = 2004
}


@INPROCEEDINGS{Godsill_Vermaak_2005,
AUTHOR = {S. J.
Godsill and J. Vermaak},
TITLE = {VARIABLE RATE PARTICLE FILTERS FOR
TRACKING APPLICATIONS},
BOOKTITLE = {Proc. IEEE Stat. Sig. Proc.,
Bordeaux},
YEAR = 2005,
MONTH = {July},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/05/multirate_ssp.pdf}
}


@INPROCEEDINGS{Gilholm_Godsill_Maskell_Salmond_2005,
AUTHOR = {K.
Gilholm and S.J. Godsill and S. Maskell and D. Salmond},
TITLE = {Poisson models for extended target and group tracking},
BOOKTITLE = {Proc. SPIE: Signal and Data Processing of Small Targets},
NUMBER = {},
YEAR = {2005},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/05/spie_poisson_05.pdf}
}


@INPROCEEDINGS{Wolfe_Godsill_2005,
AUTHOR = {P.J. Wolfe and S. J.
Godsill},
TITLE = {Interpolation of missing data values for audio
signal restoration using a {G}abor regression model },
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/05/wolfe_icassp_05.pdf},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = 2005,
MONTH = MAR
}


@INPROCEEDINGS{Wolfe_Godsill_2003,
AUTHOR = {P.J. Wolfe and S. J.
Godsill},
TITLE = {A {G}abor regression scheme for audio signal
analysis},
BOOKTITLE = {Proc. of
{IEEE} Workshop on Applications of Signal Processing to Audio and
Acoustics},
YEAR = 2003,
MONTH = OCT
}


@INPROCEEDINGS{Fevotte_Godsill_2005,
AUTHOR = {C. F\'{e}votte and S.
J. Godsill},
TITLE = {A {B}ayesian approach to time-frequency based
blind source separation},
BOOKTITLE = {Proc. of
{IEEE} Workshop on Applications of Signal Processing to Audio and
Acoustics},
YEAR = 2005,
MONTH = OCT,
URL = {http://www-sigproc.eng.cam.ac.uk/~cf269/Proceedings/waspaa05_2.pdf},
URL = {http://www-sigproc.eng.cam.ac.uk/~cf269/waspaa05_2/sound_files.pdf}
}


@INPROCEEDINGS{Cemgil_Godsill_2005,
AUTHOR = {Cemgil, A. T. and Godsill, S. J.},
TITLE = {Efficient Variational Inference for the Dynamic Harmonic Model},
BOOKTITLE = {Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {October},
OPTORGANIZATION = {},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {http://www-sigproc.eng.cam.ac.uk/~atc27/papers/cemgil-waspaa05.pdf},
ABSTRACT = {In this paper, we develop a class of probability models that are
potentially useful for various music applications such as polyphonic
transcription, source separation, restoration or denoising.  This
class unifies and extends several models such as sinusoidal and
harmonic models, additive synthesis model, Gabor regression and
probabilistic phase vocoder. We overcome computational
intractability issues by introducing structured variational
(mean-field) approximations that lead to efficient local message
passing algorithms.}
}


@INPROCEEDINGS{Ng_Li_Godsill_Vermaak_2005a,
AUTHOR = {W. Ng and J.F. Li and S.J. Godsill and J. Vermaak},
TITLE = {Tracking variable number of targets using Sequential {M}onte
{C}arlo Methods},
BOOKTITLE = {Proc. IEEE Stat. Sig. Proc.},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/ssp2005.pdf}
}


@INPROCEEDINGS{Ng_Li_Godsill_Vermaak_2005b,
AUTHOR = {W. Ng and J.F.
Li and S.J. Godsill and J. Vermaak},
TITLE = {A review of recent
results in multiple target tracking},
BOOKTITLE = {International
Symposium on Image and Signal Processing and Analysis},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/ssp2005.pdf}
}


@INPROCEEDINGS{Li_Ng_Godsill_Vermaak_2005,
AUTHOR = {J.F. Li and W. Ng
and S.J. Godsill and J. Vermaak},
TITLE = {Online Multitarget
Detection and Tracking Using Sequential {M}onte {C}arlo Methods},
BOOKTITLE = {Eighth International Conference on Information Fusion},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/fusion2005.pdf}
}


@INPROCEEDINGS{Ng_Li_Godsill_Vermaak_2005c,
AUTHOR = {W. Ng and J.F.
Li and S.J. Godsill and J. Vermaak},
TITLE = {Tracking variable number
of targets using Sequential {M}onte {C}arlo Method},
BOOKTITLE = {13th
{E}uropean {S}ignal {P}rocessing {C}onference},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/eusipco2005.pdf}
}


@INPROCEEDINGS{Ng_Li_Godsill_Vermaak_2005d,
AUTHOR = {W. Ng and J.F.
Li and S.J. Godsill and J. Vermaak},
TITLE = {A hybrid approach for
online joint detection and tracking for multiple targets},
BOOKTITLE = {IEEE Aerospace Conference},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/aes2005.pdf}
}


@INPROCEEDINGS{Ng_Li_Godsill_Vermaak_2005e,
AUTHOR = {W. Ng and J.F.
Li and S.J. Godsill and J. Vermaak},
TITLE = {Multiple target tracking
using a new soft-gating approach and sequential {M}onte {C}arlo
methods},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/underconstruction.pdf}
}


@INPROCEEDINGS{Lin_Godsill_2005,
AUTHOR = {H. Lin and S.J. Godsill},
TITLE = {The Multi-Channel {AR} Model For Real-time Audio Restoration},
BOOKTITLE = {Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {October},
OPTORGANIZATION = {},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {}
}


@INPROCEEDINGS{Godsill_Davy_2005,
AUTHOR = {S.J. Godsill and M. Davy},
TITLE = {BAYESIAN COMPUTATIONAL MODELS FOR INHARMONICITY IN
MUSICAL INSTRUMENTS},
BOOKTITLE = {Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {October},
OPTORGANIZATION = {},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/05/godsill_mohonk_05.pdf}
}


@INPROCEEDINGS{Godsill_Li_Ng_2005,
AUTHOR = {S.J. Godsill and J.F.
Li and W. Ng},
TITLE = {MULTIPLE AND EXTENDED OBJECT TRACKING WITH
{P}OISSON SPATIAL PROCESSES AND VARIABLE RATE FILTERS},
BOOKTITLE = {IEEE CAMSAP},
YEAR = {2005},
URL = {http://www-sigproc.eng.cam.ac.uk/~kfn20/papers/CAMSAP2005.pdf}
}


@INPROCEEDINGS{Ng_Pang_Li_Godsill_2006,
AUTHOR = {W. Ng and S.K.
Pang and J.F. Li and S.J. Godsill},
TITLE = {Efficient variable rate
particle filters for tracking of manoeuvring targets using an
{MRF}-based motion model},
BOOKTITLE = {EUSIPCO},
YEAR = {2006}
}


@INPROCEEDINGS{Fevotte_etal_04,
AUTHOR = {C.~F\'{e}votte and
S.~J.~Godsill and P.~J.~Wolfe},
TITLE = {{B}ayesian approach for blind
separation of underdetermined mixtures of sparse sources},
BOOKTITLE = {Proc. 5th International Conference on Independent
Component Analysis and Blind Source Separation (ICA'04)},
MONTH = SEP,
YEAR = {2004},
PAGES = {}
}


@INPROCEEDINGS{Fevotte_Godsill_2005,
AUTHOR = {C.~F\'{e}votte and
S.~J.~Godsill},
TITLE = {A {B}ayesian approach to time-frequency based
blind source separation},
BOOKTITLE = {Proc.~IEEE Workshop on
Applications of Signal Processing to Audio and Acoustics
(WASPAA'05)},
MONTH = OCT,
YEAR = {2005},
PAGES = {}
}


@INPROCEEDINGS{Fevotte_Godsill_2006,
AUTHOR = {C.~F\'evotte and
S.~J.~Godsill},
TITLE = {Blind separation of sparse sources using
{J}effrey's inverse prior and the {EM} algorithm},
BOOKTITLE = {Proc.~6th International Conference on Independent
Component Analysis and Blind Source Separation (ICA'06)},
MONTH = {Mar.},
YEAR = {2006},
PAGES = {}
}


@INPROCEEDINGS{Fevotte_et_al_2006,
AUTHOR = {C.~F\'evotte and
L.~Daudet and S.~J.~Godsill and B.~Torr\'esani},
TITLE = {SPARSE
REGRESSION WITH STRUCTURED PRIORS: APPLICATION TO AUDIO DENOISING},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
MONTH = {May},
YEAR = {2006},
PAGES = {}
}


@INPROCEEDINGS{Fevotte_Godsill_2006c,
AUTHOR = {C.~F\'evotte and S.~J.~Godsill},
TITLE = {Blind separation of
sparse sources using {J}effrey's inverse prior and the {EM}
algorithm},
BOOKTITLE = {Proc.~6th International Conference on
Independent Component Analysis and Blind Source Separation
(ICA'06)},
MONTH = {Mar.},
YEAR = {2006},
PAGES = {}
}


@INPROCEEDINGS{Godsill_2007,
AUTHOR = {S.J. Godsill},
BOOKTITLE = {ESAIM: PROCEEDINGS of Oxford Workshop on Particle
Filtering},
YEAR = 2007,
TITLE = {PARTICLE FILTERS FOR CONTINUOUS-TIME
JUMP MODELS IN TRACKING APPLICATIONS},
NOTE = {(To Appear)}
}


@INPROCEEDINGS{Godsill_Yang_2006,
AUTHOR = {S.J. Godsill and L. Yang},
TITLE = {BAYESIAN INFERENCE FOR CONTINUOUS-TIME {AR} MODELS DRIVEN BY
NON-{G}AUSSIAN L\'{E}VY PROCESSES},
BOOKTITLE = {Proc. IEEE International Conference on Acoustics,
Speech and Signal Processing},
MONTH = {May},
YEAR = {2006},
PAGES = {}
}


@INPROCEEDINGS{cemgil:waspaa05,
AUTHOR = {Cemgil, A. T. and Godsill, S. J.},
TITLE = {{E}fficient {V}ariational {I}nference for the {D}ynamic {H}armonic {M}odel},
BOOKTITLE = {Proc. of IEEE Workshop on Applications of Signal Processing to Audio and Acoustics},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {October},
OPTORGANIZATION = {},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/cemgil-godsill-dhm.pdf},
ABSTRACT = {In this paper, we develop a class of probability models that are
potentially useful for various music applications such as polyphonic
transcription, source separation, restoration or denoising.  This
class unifies and extends several models such as sinusoidal and
harmonic models, additive synthesis model, Gabor regression and
probabilistic phase vocoder. We overcome computational
intractability issues by introducing structured variational
(mean-field) approximations that lead to efficient local message
passing algorithms.}
}


@INPROCEEDINGS{cemgil-eusipco1,
AUTHOR = {Cemgil, A. T. and Godsill, S. J.},
TITLE = {Probabilistic {P}hase {V}ocoder and its application to {I}nterpolation of {M}issing {V}alues in {A}udio {S}ignals},
BOOKTITLE = {13th {E}uropean {S}ignal {P}rocessing {C}onference},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
OPTMONTH = {},
ORGANIZATION = {{EURASIP}},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/cemgil-godsill-em-restore-eusipco.pdf},
URL = {http://www-sigproc.eng.cam.ac.uk/~atc27/em-restore/},
ABSTRACT = {We formulate the phase vocoder -- an audio synthesis method very
closely related to inverse short time Fourier Transform synthesis --
as a Gaussian state space model and demonstrate simulation results
on interpolation of missing values. The audio signal is modelled as
a superposition of quasi-sinusoidal signals generated by a linear
dynamical system.  The advantage of our generative'' perspective
is that it allows a full Bayesian treatment of the problem; e.g. one
can perform the analysis while arbitrary chunks of sample values are
missing or model parameters are unknown.  To perform audio
restoration, we derive an  expectation-maximisation (EM) algorithm
that infers the expectations of missing samples and maximum
a-posteriori model parameters.  We demonstrate the validity of our
approach on a set of challenging real audio examples and compare to
existing methods.}
}


@INPROCEEDINGS{cemgil-eusipco2,
AUTHOR = {Cemgil, A. T. and Fevotte, C. and Godsill, S. J.},
TITLE = {Blind {S}eparation of {S}parse {S}ources using {V}ariational {EM}},
BOOKTITLE = {13th {E}uropean {S}ignal {P}rocessing {C}onference},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2005},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
OPTMONTH = {},
ORGANIZATION = {{EURASIP}},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/cemgil-fevotte-godsill-bss-eusipco05.pdf},
URL = {http://www-sigproc.eng.cam.ac.uk/~cf269/eusipco05/sound_files.html},
ABSTRACT = {In this paper, we tackle the general linear instantaneous model
(possibly underdetermined and noisy) using the assumption of
\textit{sparsity} of the sources on a given dictionary. We model the
sparsity of expansion coefficients with a Student \emph{t} prior.
The conjugate-exponential characterisation of the \emph{t}
distribution as an infinite mixture of scaled Gaussians enables us
to derive an efficient variational expectation maximisation
algorithm (V-EM). The resulting deterministic algorithm has superior
properties in terms of computation time and achieves a separation
performance comparable in quality to alternative methods based on
Markov Chain Monte Carlo (MCMC).}
}


@INPROCEEDINGS{whiteley-ismir06,
AUTHOR = {Whiteley, N. and Cemgil, A. T. and Godsill, S. J.},
TITLE = {Bayesian Modelling of Temporal Structure in Musical Audio},
BOOKTITLE = {Proceedings of International Conference on Music Information Retrieval},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2006},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
OPTMONTH = {},
OPTORGANIZATION = {},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/whiteley-cemgil-godsill-ismir06.pdf},
ABSTRACT = {This paper presents a probabilistic model of temporal structure in music
which allows joint inference of tempo, meter and rhythmic pattern.
The framework of the model naturally quantifies these three musical
concepts in terms of hidden state-variables, allowing resolution of
otherwise apparent ambiguities in musical structure. At the heart of
the system is a probabilistic model of a hypothetical bar-pointer'
which maps an input signal to one cycle of a latent, periodic
rhythmical pattern. The system flexibly accommodates different input
signals via two observation models: a Poisson points model for use
with MIDI onset data and a Gaussian process model for use with raw
audio signals. The discrete state-space permits exact computation of
posterior probability distributions for the quantities of interest.
Results are presented for both observation models, demonstrating the
ability of the system to correctly detect changes in rhythmic
pattern and meter, whilst tracking tempo.}
}


@INPROCEEDINGS{Fallon_Godsill_Blake_2006,
AUTHOR = {M. Fallon and S.
Godsill and A. Blake},
TITLE = {Joint Acoustic Source Location and
Orientation Estimation using Sequential Monte Carlo},
BOOKTITLE = {Proc. of the 9th Int. Conference on Digital Audio Effects
MONTH = SEP,
YEAR = 2006
}


@INPROCEEDINGS{Lin_Godsill_2006,
AUTHOR = {H. Lin and S. Godsill},
TITLE = {Real-Time Bayesian GSM Buzz Removal},
BOOKTITLE = {Proc. of the
9th Int. Conference on Digital Audio Effects (DAFx'06), Montreal,
MONTH = SEP,
YEAR = 2006
}


@ARTICLE{Yoon_Godsill_2006,
AUTHOR = {J.W. Yoon and S.J. Godsill},
TITLE = {Bayesian Inference for
Multidimensional NMR image reconstruction},
BOOKTITLE = {European
Signal Processing Conference (EUSIPCO), Florence, Italy},
MONTH = SEP,
YEAR = 2006
}


@INPROCEEDINGS{godsill-eusipco07,
AUTHOR = {S.J. Godsill and A.T. Cemgil and C. Fevotte and P.J. Wolfe},
TITLE = {{B}ayesian computational methods for sparse audio and music
processing},
BOOKTITLE = {15th {E}uropean {S}ignal {P}rocessing {C}onference},
OPTCROSSREF = {},
OPTKEY = {},
OPTPAGES = {},
YEAR = {2007},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
OPTMONTH = {},
ORGANIZATION = {{EURASIP}},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/godsill-cemgil-fevotte-wolfe-eusipco07.pdf},
ABSTRACT = {In this paper we provide an overview of some recently developed
Bayesian models and algorithms for estimation of sparse signals. The
models encapsulate the sparseness inherent in audio and musical
signals through structured sparsity priors on coefficients in the
model. Markov chain Monte Carlo (MCMC) and variational methods are
described for inference about the parameters and coefficients of
these models, and brief simulation examples are given.}
}


@INPROCEEDINGS{whiteley-icassp07,
AUTHOR = {Whiteley, N. and Cemgil, A. T. and Godsill, S. J.},
TITLE = {{S}equential {I}nference of {R}hythmic {S}tructure in {M}usical {A}udio},
BOOKTITLE = {Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP 07)},
OPTCROSSREF = {},
OPTKEY = {},
PAGES = {1321-1324},
YEAR = {2007},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {April},
ORGANIZATION = {IEEE},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/whiteley-cemgil-godsill-icassp07.pdf},
ABSTRACT = {This paper presents a framework for the modelling of temporal
characteristics of musical signals and an approximate, sequential
Monte Carlo inference scheme which yields estimates of tempo and
rhythmic pattern from onset-time data. These two features are
quantified through the construction of a probabilistic dynamical
model of a hidden bar-pointer and a Cox process observation model.
The capabilities of the system are demonstrated by tracking the
tempo of a 2 against 3 polyrhythm and detecting a switch in rhythm
in a MIDI performance.}
}


@INPROCEEDINGS{whiteley-icassp07,
AUTHOR = {Whiteley, N. and Cemgil, A. T. and Godsill, S. J.},
TITLE = {{S}equential {I}nference of {R}hythmic {S}tructure in {M}usical {A}udio},
BOOKTITLE = {Proc. of IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP 07)},
OPTCROSSREF = {},
OPTKEY = {},
PAGES = {1321-1324},
YEAR = {2007},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {April},
ORGANIZATION = {IEEE},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {},
PDF = {papers/whiteley-cemgil-godsill-icassp07.pdf},
ABSTRACT = {This paper presents a framework for the modelling of temporal
characteristics of musical signals and an approximate, sequential
Monte Carlo inference scheme which yields estimates of tempo and
rhythmic pattern from onset-time data. These two features are
quantified through the construction of a probabilistic dynamical
model of a hidden bar-pointer and a Cox process observation model.
The capabilities of the system are demonstrated by tracking the
tempo of a 2 against 3 polyrhythm and detecting a switch in rhythm
in a MIDI performance.}
}


@INPROCEEDINGS{whiteley-ispa07,
AUTHOR = {Whiteley, N. and Singh, S. and Godsill, S. J.},
TITLE = {Auxiliary Particle Implementation of the {P}robability {H}ypothesis {D}ensity Filter},
BOOKTITLE = {Proc. of 5th International Symposium on
Image and Signal Processing and Analysis},
OPTCROSSREF = {},
OPTKEY = {},
PAGES = {},
YEAR = {2007},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = SEP,
ORGANIZATION = {IEEE},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {}
}


@INPROCEEDINGS{Pang_Li_Godsill,
AUTHOR = {S.K. Pang and J-F. Li and S.J. Godsill},
TITLE = {Models and Algorithms for Detection and Tracking of Coordinated Groups},
BOOKTITLE = {Proc. of 5th International Symposium on
Image and Signal Processing and Analysis},
OPTCROSSREF = {},
OPTKEY = {},
PAGES = {},
YEAR = {2007},
OPTEDITOR = {},
OPTVOLUME = {},
OPTNUMBER = {},
OPTSERIES = {},
MONTH = {Sep},
ORGANIZATION = {IEEE},
OPTPUBLISHER = {},
OPTNOTE = {},
OPTANNOTE = {}
}


@INPROCEEDINGS{peeling07ISMIR,
AUTHOR = {P. H. Peeling and A. T. Cemgil and S. J. Godsill},
TITLE = {A Probabilistic Framework for Matching Music Representations},
BOOKTITLE = {Proc. ISMIR},
YEAR = 2007,
URL = {http://www-sigproc.eng.cam.ac.uk/~php23/publications/ISMIR/}
}


@INPROCEEDINGS{cemgil07WASPAA,
AUTHOR = {A. T. Cemgil and P. H. Peeling and O. Dikmen and S. J. Godsill},
TITLE = {Prior structures for time-frequency energy distributions},
BOOKTITLE = {Proc. IEEE WASPAA},
YEAR = 2007,
URL = {http://www-sigproc.eng.cam.ac.uk/~php23/publications/WASPAA/}
}


@INPROCEEDINGS{Fallon_Godsill_2007,
AUTHOR = {M. Fallon and S.J.
Godsill},
TITLE = {MULTI TARGET ACOUSTIC SOURCE TRACKING USING TRACK
BEFORE DETECT},
BOOKTITLE = {Proc. IEEE WASPAA},
MONTH = OCT,
YEAR = 2007,
NOTE = {}
}


@INPROCEEDINGS{Whiteley_Johanssen_Godsill_2007,
AUTHOR = {N. Whiteley
and A.M. Johanssen and S.J. Godsill},
TITLE = {EFFICIENT MONTE CARLO
FILTERING FOR DISCRETELY OBSERVED JUMPING PROCESSES},
BOOKTITLE = {Proc. IEEE SSP},
YEAR = 2007
}


@INPROCEEDINGS{Yang_Godsill_2007,
AUTHOR = {G. Yang and S.J. Godsill},
TITLE = {BAYESIAN INFERENCE FOR CONTINUOUS-TIME {ARMA} MODELS DRIVEN
BY JUMP diffusions},
BOOKTITLE = {Proc. IEEE SSP},
YEAR = 2007
}


@ARTICLE{Godsill_Rayner_1995a,
AUTHOR = {S.  J.  Godsill and P.  J.  W.  Rayner},
TITLE = {A {B}ayesian Approach to the Restoration of Degraded Audio Signals},
JOURNAL = {IEEE Trans. on
Speech and Audio Processing},
YEAR = 1995,
MONTH = JUL,
VOLUME = 3,
NUMBER = 4,
PAGES = {267--278},
ABSTRACT = {
In this paper we derive the {\em a posteriori} probability for the
location of bursts of noise additively superimposed on a Gaussian AR
process. The theory is developed to give a sequentially based
restoration algorithm suitable for real-time applications. The
algorithm is particularly appropriate for digital audio restoration,
where clicks and scratches may be modelled as additive bursts of
noise. Experiments are carried out on both real audio data and
synthetic AR processes and significant improvements are demonstrated
over existing restoration techniques.
}
}


@ARTICLE{Godsill_Rayner_1997,
AUTHOR = {S. J.  Godsill and P. J. W.  Rayner},
TITLE = {Robust reconstruction and analysis of autoregressive signals in impulsive noise using the {G}ibbs Sampler},
JOURNAL = {IEEE Trans. on
Speech and Audio Processing},
VOLUME = {6},
NUMBER = {4},
YEAR = 1998,
MONTH = JUL,
PAGES = {352--372},
ABSTRACT = {
Modelling and reconstruction methods are presented for noise
reduction of autocorrelated signals in non-Gaussian, impulsive noise
environments. A Bayesian probabilistic framework is adopted and
Markov chain Monte Carlo methods are developed for detection and
correction of impulses. Individual noise sources are modelled as
Gaussian with unknown scale (variance), allowing for robustness to
heavy-tailed' impulse distributions, while the underlying signal is
modelled as autoregressive (AR). Results are presented for both
artificial and real data from voice and music recordings and
comparisons are made with existing techniques. The new techniques
are found to give improved detection and elimination of impulses in
adverse noise conditions at the expense of some extra computational
complexity.
},
PDF = {http://www.com-serv.eng.cam.ac.uk/~sjg/papers/95/impulse.ps.gz}
}


@ARTICLE{Rajan_Rayner_Godsill_1997,
AUTHOR = {J. J. Rajan  and P. J. W. Rayner and S. J. Godsill},
TITLE = { A {B}ayesian Approach to Parameter Estimation and
Interpolation of Time-Varying Autoregressive Processes using the
{G}ibbs Sampler },
JOURNAL = {IEE Proc. Vision, Image and Signal Processing},
YEAR = 1997,
VOLUME = 144,
NUMBER = 4,
MONTH = AUG,
PS = {http://www-com-serv.eng.cam.ac.uk/~sjg/papers/97/gibbsiee.ps.gz},
ABSTRACT = {
A non-stationary time series is one in which the statistics of the
process are a function of time; this time dependency makes it
impossible to utilize standard analytically defined statistical
estimators to parameterize the process.  In order to overcome this
difficulty the time series is considered within a finite time
interval and is modelled as a time varying autoregressive (AR)
process.  The AR coefficients that characterize this model are
functions of time that may be represented by a family of basis
vectors.  The corresponding basis coefficients are invariant over
the time window and have stationary statistical properties.
}
}


@ARTICLE{Godsill_1996b,
AUTHOR = {S.  J.  Godsill},
TITLE = {{B}ayesian enhancement of speech and audio signals
which can be modelled as {ARMA} processes},
JOURNAL = {International Statistical Review},
YEAR = 1997,
VOLUME = 65,
NUMBER = 1,
PAGES = {1--21},
ABSTRACT = {
In application areas which involve digitised speech and audio signals,
such as coding, digital remastering of old recordings and
recognition of speech, it is often desirable to reduce the effects
of noise with the aim of enhancing intelligibility and perceived
sound quality.
We consider the case where noise
sources contain non-Gaussian, impulsive elements superimposed upon a
continuous Gaussian background. Such a situation arises in areas
such as communications channels, telephony and gramophone recordings
where impulsive effects might be caused by electromagnetic
interference (lightning strikes), electrical switching noise or
defects in recording media, while electrical circuit noise or the
combined effect of many distant atmospheric events lead to a
continuous Gaussian component.

In this paper we discuss the background to this type of noise
degradation and describe briefly some existing statistical
techniques for noise reduction. We propose new methods for
enhancement based upon Markov chain Monte Carlo (MCMC) simulation.
Signals are modelled as autoregressive moving-average (ARMA), while
noise sources are treated as discrete and continuous mixtures of
Gaussian distributions. Results are presented for both real and
artificially corrupted data sequences, illustrating the potential of
the new methods.
},
GZ = {http://www-com-serv.eng.cam.ac.uk/~sjg/papers/97/isr97.ps.gz}
}


@ARTICLE{Doucet_Godsill_Andrieu_1999,
TITLE = {On sequential {M}onte {C}arlo sampling methods for {B}ayesian filtering},
AUTHOR = {A. Doucet and S. J. Godsill and C. Andrieu},
JOURNAL = {Statistics and Computing},
YEAR = 2000,
VOLUME = {10},
PAGES = {197--208},
ABSTRACT = {},
URL = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/99/statcomp_final.ps}
}


@ARTICLE{Troughton_Godsill_1999a,
TITLE = {{MCMC} methods for restoration of nonlinearly distorted
autoregressive signals},
AUTHOR = {P.T. Troughton and S.J. Godsill},
JOURNAL = {Signal Processing},
YEAR = 2001,
VOLUME = 81,
NUMBER = 1,
PAGES = {83--97},
ABSTRACT = {We approach the problem of restoring distorted autoregressive (AR)
signals by using a cascade model, in which the observed signal is
modelled as the output of a nonlinear AR (NAR) process excited by
the linear AR signal we are attempting to recover.

The Volterra expansion of the NAR model has a very large number of
possible terms even when truncated at fairly small maximum
polynomial degrees and lags. We address the problem of subset
selection and uncertainty in the nonlinear stage and model order
uncertainty in the linear stage through a hierarchical Bayesian
approach. A Markov chain Monte Carlo (MCMC) approach is used for
implementation, with \rj\ moves for the linear model order and a
rapidly mixing Gibbs sampler for subset selection in the nonlinear
model stage, exploiting the partially analytic properties of the
model.

We demonstrate the method using both synthetic AR data and an audio
extract, and extend the approach to process a long distorted audio
time series, for which the source model cannot be considered to be
time-invariant},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/99/sparnar.ps}
}


@ARTICLE{Godsill_1997b,
AUTHOR = {S. J. Godsill},
TITLE = {On the Relationship Between {M}arkov chain {M}onte
{C}arlo Methods for Model Uncertainty},
YEAR = 2001,
JOURNAL = {J. Comp. Graph. Stats.},
ABSTRACT = {
We discuss relationships between the existing methods for MCMC
exploration of model spaces, including the reversible jump sampler
of Green (1995), the model composition' approach of Carlin and Chib
(1995), the MC$^3$ techniques of Madigan and Raftery (1995) and MCMC
methods for variable selection such as George and McCulloch (1993),
Kuo and Mallick (1997) and Geweke (1996). We link these different
methods together through a composite model space similar to that
used by Carlin and Chib in which a model of constant dimensionality
is created by considering the product space of parameters from all
possible models within the candidate set and the model indexing
variable. In the examples given in their paper, Carlin and Chib
apply a straightforward Gibbs sampler to the composite space which
renders the method impracticable for comparison between more than a
small handful of models. We show that the other methods of MCMC
model selection can be obtained by applying different forms of MCMC
sampling to the composite space. The results shed some light upon
the issues of pseudo-prior' selection in the case of the Carlin and
Chib sampler and choice of proposal distribution in the case
of Green's reversible jump method. Furthermore, we propose efficient reversible jump proposal schemes which
take advantage of any analytic structure that may be present in the
model. The method is compared with a standard reversible jump scheme
for the problem of model order uncertainty in  autoregressive time
series},
VOLUME = 10,
NUMBER = 2,
PAGES = {230--248},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/model_jcgs.pdf}
}


@ARTICLE{Godsill_Doucet_West_00c,
AUTHOR = {S. J. Godsill and A Doucet and M West},
JOURNAL = {Ann. Inst. Stat. Math.},
VOLUME = 53,
NUMBER = 1,
TITLE = {Maximum \emph{a posteriori} sequence estimation using {M}onte {C}arlo particle filters},
MONTH = MAR,
PAGES = {82--96},
YEAR = {2001},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/map_sequence.ps}
}


@ARTICLE{Doucet_Godsill_Robert_2000,
AUTHOR = {A. Doucet and S. J. Godsill and C. P. Robert},
TITLE = {Marginal Maximum A Posteriori Estimation using
{MCMC}},
JOURNAL = {Statistics and Computing},
VOLUME = 12,
PAGES = {77--84},
ABSTRACT = {Markov chain Monte Carlo (MCMC) methods, while
facilitating the solution of many complex problems
in Bayesian inference, are not currently well
adapted to the problem of marginal \textit{maximum a
posteriori} (MMAP) estimation, especially when the
number of parameters is large. We present here a
simple and novel Markov Chain Monte Carlo (MCMC)
strategy, called \textit{ State-Augmentation for
Marginal Estimation} (SAME), which leads to MMAP
estimates for Bayesian models. We illustrate the
simplicity and utility of the approach for missing
data interpolation in autoregressive time series and
blind deconvolution of impulsive processes.},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/same.ps},
YEAR = 2002
}


@ARTICLE{Wolfe_Godsill_2001a,
AUTHOR = {P. J. Wolfe and S. J.
Godsill},
TITLE = {Perceptually motivated approaches to music
restoration},
EDITOR = {M. Dörfler and H. G. Feichtinger },
NOTE = {Special issue: Music and Mathematics},
JOURNAL = {Journal of New
Music Research},
VOLUME = 30,
NUMBER = 1,
YEAR = 2001,
PS = {http://www-sigproc.eng.cam.ac.uk/~{}sjg/papers/01/jnmr.ps},
PAGES = {83--92}
}


@ARTICLE{Godsill_Wolfe_Fong_2001,
AUTHOR = {S. J.
Godsill and  P. J. Wolfe and W. N. W. Fong},
TITLE = {Statistical
model-based approaches to audio restoration and analysis},
EDITOR = {S. Canazza and A. Vidolin},
JOURNAL = {Journal of New Music
Research},
YEAR = 2001,
NOTE = { Special Issue: Conservation,
Restoration and Archiving of Electroacoustic Music},
VOLUME = 30,
NUMBER = 4,
PAGES = {323--338}
}


@ARTICLE{fong_godsill_doucet_west_2001,
AUTHOR = {W. Fong and S. J. Godsill and A. Doucet and M. West},
TITLE = {Monte {C}arlo Smoothing with application to speech
enhancement},
JOURNAL = {IEEE Trans. on Signal
Processing},
NOTE = {Special issue on Monte
Carlo Methods},
YEAR = 2002,
MONTH = FEB,
VOLUME = 50,
NUMBER = 2,
PAGES = {438--449},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/fong_smoother_IEEE_SP.ps}
}


@ARTICLE{kokaram_godsill_2001,
AUTHOR = {A.C. Kokaram and S.J.
Godsill},
TITLE = {{MCMC} for joint noise reduction and missing data
JOURNAL = {IEEE Trans. on Signal
Processing},
NOTE = {Special
issue on Monte Carlo Methods},
YEAR = 2002,
MONTH = FEB,
VOLUME = 50,
NUMBER = 2,
PAGES = {189--205}
}


@ARTICLE{Vermaak_Andrieu_Doucet_Godsill_2000a,
AUTHOR = {J. Vermaak and C. Andrieu and A. Doucet and S. J. Godsill},
TITLE = {Reversible Jump {M}arkov chain {M}onte {C}arlo strategies for {B}ayesian Model Selection in Autoregressive Processes},
YEAR = 2004,
JOURNAL = {J. Time Series Anal.},
GZ = {http://www-svr.eng.cam.ac.uk/~jv211/papers/report_arms.ps.gz},
MONTH = NOV,
PAGES = {785--945},
VOLUME = 25,
NUMBER = 6
}


@ARTICLE{Vermaak_Andrieu_Doucet_Godsill_1999a,
AUTHOR = {J. Vermaak and C. Andrieu and A. Doucet and S. J. Godsill},
TITLE = {Particle Methods for {B}ayesian Modelling and Enhancement of Speech Signals},
JOURNAL = {IEEE Trans. on
Speech and Audio Processing},
VOLUME = 10,
NUMBER = 3,
PAGES = {173--185},
YEAR = 2002,
GZ = {http://www-svr.eng.cam.ac.uk/~jv211/papers/paper_tvar_seq.ps.gz}
}


@ARTICLE{Godsill_Doucet_West_00d,
AUTHOR = {S. J. Godsill and A. Doucet and M. West},
JOURNAL = {J. Am. Statist. Assoc.},
TITLE = {{M}onte {C}arlo smoothing for non-linear time series},
YEAR = {2004},
VOLUME = 99,
NUMBER = 465,
PAGES = {156--168},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/00/smoother.ps}
}


@ARTICLE{Wolfe_Godsill_2003a,
AUTHOR = {P. J. Wolfe and S. J. Godsill},
JOURNAL = {EURASIP Journal on Appl. Sig. Processing},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/03/wolfe_jasp_03.pdf},
TITLE = {Efficient Alternatives to the {E}phraim and {M}alah suppression rule for audio signal enhancement},
PAGES = {1043--1051},
VOLUME = 10,
NUMBER = 1,
YEAR = {2003}
}


@ARTICLE{Wolfe_Godsill_Ng_2003,
AUTHOR = {P. J. Wolfe and S. J. Godsill and W.J. Ng},
JOURNAL = {Journal of the Royal Statistical Society,
Series B},
VOLUME = 66,
NUMBER = 3,
PAGES = {575--589},
NOTE = {Read paper (with discussion).},
TITLE = {Bayesian variable selection and regularisation for time-frequency surface estimation},
YEAR = {2004}
}


@ARTICLE{Vermaak_Godsill_Perez_2004,
AUTHOR = {J. Vermaak and S. Godsill and P. Perez},
TITLE = {Monte {C}arlo filtering for multi-target tracking and data association},
YEAR = 2005,
JOURNAL = {IEEE Tr. Aerospace and Electronic Systems},
MONTH = JAN,
VOLUME = 41,
NUMBER = 1,
PAGES = {309--332},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/04/multi_object_revI.pdf}
}


@ARTICLE{Vermaak_Ikoma_Godsill_2005,
AUTHOR = {J. Vermaak and N.
Ikoma and S.J. Godsill},
TITLE = {Sequential {M}onte {C}arlo framework
for extended object tracking},
Navig.},
VOLUME = {152},
NUMBER = {5},
YEAR = {2005},
MONTH = OCT,
PAGES = {353--363},
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/05/iee_track.pdf}
}


@ARTICLE{Fevotte_Godsill_2006,
AUTHOR = {C. F\'{e}votte and S.J.
Godsill},
TITLE = {A {B}ayesian approach for blind separation of
sparse sources},
URL = {http://www-sigproc.eng.cam.ac.uk/~cf269/Journals/ieee_sap06.pdf},
JOURNAL = {IEEE Trans. on
Speech and Audio Processing},
YEAR = 2006
}


@ARTICLE{Yoon_Godsill_Kupce_Freeman_2005,
AUTHOR = {Ji Won Yoon and
Simon Godsill and Eriks Kupce and Ray Freeman},
TITLE = {Deterministic
and statistical methods for reconstructing multidimensional NMR
spectra},
JOURNAL = {Magnetic Resonance in Chemistry},
MONTH = MAR,
YEAR = {2006}
}


@ARTICLE{Lombardi_Godsill_2005,
YEAR = 2006,
AUTHOR = {M.
Lombardi and S.J. Godsill},
TITLE = {On-line {B}ayesian Estimation of
{AR} Signals in Symmetric alpha-Stable Noise},
URL = {http://ideas.repec.org/p/fir/econom/wp2004_05.html},
JOURNAL = {IEEE Trans. on Signal
Processing}
}


@ARTICLE{Fevotte_Godsill_2006a,
AUTHOR = {C.
F\'{e}votte and S.J. Godsill},
TITLE = {A {B}ayesian approach for
blind separation of sparse sources},
JOURNAL = {IEEE Trans. on
Speech and Audio Processing},
VOLUME = 14,
NUMBER = 6,
YEAR = 2006,
PAGE = {2174--2188}
}


@ARTICLE{Fevotte_Godsill_2006b,
AUTHOR = {C.~F\'{e}votte and
S.~Godsill},
TITLE = {Sparse linear regression in unions of bases via
{B}ayesian variable selection},
JOURNAL = {IEEE Signal Processing
Letters},
VOLUME = {13},
NUMBER = {7},
PAGES = {441--444},
MONTH = JUL,
YEAR = {2006},
PDF = {http://persos.mist-technologies.com/.cfevotte/}
}


@ARTICLE{cemgil:bss06,
AUTHOR = {Cemgil, A. T. and Godsill, S. J. and Fevotte, C.},
TITLE = {Variational and {S}tochastic {I}nference for {B}ayesian {S}ource {S}eparation},
JOURNAL = {Digital Signal Processing},
YEAR = {2007},
PAGE = {891--913},
VOLUME = {17},
OPTNUMBER = {5},
OPTPAGES = {},
OPTMONTH = {},
OPTNOTE = {Special Issue on Bayesian Source Separation},
OPTANNOTE = {},
PDF = {papers/cemgil-fevotte-godsill-dsp-bss.pdf},
ABSTRACT = {We tackle the general linear instantaneous model
(possibly underdetermined and noisy) where we model the source prior
with a Student \emph{t} distribution. The conjugate-exponential
characterisation of the \emph{t} distribution as an infinite mixture
of scaled Gaussians enables us to do efficient inference. We study
two well known inference methods, Gibbs sampler and variational
Bayes for Bayesian source separation. We derive both techniques as
local message passing algorithms  to highlight their algorithmic
similarities and to contrast their different convergence
characteristics and computational requirements. Our simulation
results suggest that typical posterior distributions in source
separation have multiple local maxima. Therefore we propose a
hybrid approach  where we explore the state space with a Gibbs
sampler and then switch to a deterministic algorithm. This approach
seems to be able to combine the speed of the variational approach
with the robustness of the Gibbs sampler.}
}


@ARTICLE{Davy_Godsill_Idier_2006,
AUTHOR = {M. Davy and S.J. Godsill
and J. Idier},
TITLE = {Bayesian Analysis of Polyphonic Western Tonal Music},
JOURNAL = {Journal of the Acoustical Society of America},
VOLUME = 119,
NUMBER = 4,
PAGE = {2498--2517},
MONTH = {April},
YEAR = {2006}
}


@ARTICLE{Godsill_Vermaak_Ng_Li_2007,
AUTHOR = {S.J. Godsill and J.
Vermaak and K-F. Ng and J-F. Li},
TITLE = {Models and Algorithms for
Tracking of Manoeuvring Objects using Variable Rate Particle
Filters},
JOURNAL = {Proc. IEEE},
MONTH = MAY,
VOLUME = 95,
NUMBER = 5,
PAGE = {925--952},
YEAR = 2007
}


@ARTICLE{Cappe_Godsill_Moulines_2007,
AUTHOR = {O. Capp\'{e} and  S.J. Godsill and E.Moulines},
TITLE = {An
overview of existing methods and recent advances in sequential Monte
Carlo},
JOURNAL = {Proc. IEEE},
MONTH = MAY,
VOLUME = 95,
NUMBER = 5,
PAGE = {899--924},
YEAR = 2007
}


@ARTICLE{peeling07jasael,
AUTHOR = {P. H. Peeling and C. Li and S. J. Godsill},
TITLE = {Poisson point process modeling for polyphonic music transcription},
JOURNAL = {Journal of the Acoustical Society of America Express Letters},
YEAR = {2007},
VOLUME = {121},
NUMBER = {4},
PAGES = {EL168--EL175},
MONTH = {April},
URL = {http://scitation.aip.org/journals/doc/JASMAN-ft/vol_121/iss_4/EL168_1-div0.html},
PDF = {http://www-sigproc.eng.cam.ac.uk/~php23/publications/JASAEL/JAS0EL168.pdf},
PS = {http://www-sigproc.eng.cam.ac.uk/~php23/publications/JASAEL/JAS0EL168.ps.gz},
NOTE = {Reused with permission from Paul Peeling, The Journal of the Acoustical Society of America, 121, EL168 (2007). Copyright 2007, Acoustical Society of America.}
}


@ARTICLE{Ng_Li_Godsill_Pang_2007,
AUTHOR = {K-F. Ng and J-F. Li and
S.J. Godsill and S.K. Pang},
TITLE = {Multitarget Initiation, Tracking
and Termination Using Bayesian Monte Carlo Methods},
JOURNAL = {The
Computer Journal},
YEAR = {2007},
NOTE = {{\em (To
appear)\/}}
}


@ARTICLE{Ridgway_Godsill_2007,
AUTHOR = {G. Ridgway and S.J. Godsill},
TITLE = {Bayesian Image Modelling of cDNA Microarray Spots},
JOURNAL = {Signal Processing Letters},
YEAR = {2007},
NOTE = {toappear}
}


@ARTICLE{Fevotte_et_al_2007,
AUTHOR = {C. Févotte and B. Torrésani and
L. Daudet and S. J. Godsill},
TITLE = {Sparse linear regression with
structured priors and application to denoising of musical audio},
JOURNAL = {IEEE Trans. Audio, Speech and Language Processing},
NOTE = {to appear},
YEAR = 2007,
PDF = {http://www.tsi.enst.fr/~fevotte/Journals/ieee_asl_sparsereg_struc.pdf}
}


@INCOLLECTION{Godsill_Rayner_Cappe_1995,
AUTHOR = {S. J.  Godsill and P. J. W.  Rayner and O.  Capp\'{e}},
EDITOR = { M.  Kahrs and K. Brandenburg },
TITLE = {Digital Audio Restoration},
BOOKTITLE = {Applications of Digital Signal Processing
to Audio and Acoustics},
PUBLISHER = {Kluwer Academic Publishers, ISBN 0-7923-8130-0},
YEAR = 1998,
PAGES = {133-193},
ABSTRACT = {
This chapter is concerned with the application of modern signal
processing techniques to the restoration of degraded audio signals.
Although attention is focussed on gramophone recordings, film sound
tracks and tape recordings, many of the techniques discussed have
applications in other areas where degraded audio signals occur, such
as speech transmission, telephony and hearing aids.

We aim to provide a wide coverage of existing methodology while
giving insight into current areas of research and future trends.
},
GZ = {http://www-com-serv.eng.cam.ac.uk/~sjg/papers/97/chapt.ps.gz}
}


@INCOLLECTION{Godsill_Rayner_1995e,
AUTHOR = {S. J. Godsill and P. J. W. Rayner},
TITLE = {Robust Treatment of IMPULSIVE NOISE IN SPEECH AND AUDIO
SIGNALS},
BOOKTITLE = {Bayesian Robustness - proceedings of the workshop on Bayesian robustness, May 22-25, 1995, Rimini, Italy},
EDITOR = {J.O. Berger and B. Betro and E. Moreno and L.R. Pericchi and F. Ruggeri and G. Salinetti and L. Wasserman},
YEAR = 1996,
PUBLISHER = {IMS Lecture Notes - Monograph Series},
VOLUME = 29,
PAGES = {331-342},
ABSTRACT = {
Markov chain Monte Carlo methods are presented for treatment of localized, impulsive noise (outliers) in digitized
waveforms, within a Bayesian hierarchical framework. Outliers in
audio signals occur as clicks' and crackles' in degraded sound
recordings and impulsive noise in communications channels.
Sampling-based methods for detection and correction of such
artefacts are presented, in which individual noise sources are
modelled as Gaussian with unknown scale, allowing for robustness to
heavy-tailed noise distributions. Results are presented for speech
and audio signals obtained from digitized sound recordings.
},
GZ = {http://www-com-serv.eng.cam.ac.uk/~sjg/papers/96/ims96.ps.gz}
}


@INPROCEEDINGS{Troughton_Godsill_1998,
AUTHOR = {P. T. Troughton and S. J. Godsill},
TITLE = {{B}ayesian Model Selection for Linear and Non-linear Time
Series using the {G}ibbs Sampler},
YEAR = 1998,
BOOKTITLE = {Mathematics in Signal Processing {IV}},
PUBLISHER = {Oxford University Press},
ABSTRACT = {We present a stochastic simulation technique for model
selection
in time series, based on the use of indicator variables with
the Gibbs sampler within a hierarchical Bayesian framework.
As an example, the method is applied to the selection of
subset
AR models, in which only significant lags are included.
The same approach is then used to identify the structure
of a non-linear time series.
We discuss the possibility of model mixing where the model
is not well
determined by the data.}
}


@INCOLLECTION{Fitzgerald_Godsill_Kokaram_Stark_1998,
BOOKTITLE = {Bayesian Statistics VI},
EDITOR = {J.M. Bernardo and J.O. Berger and  A.P. Dawid and A.F.M. Smith},
PUBLISHER = {Oxford University Press},
AUTHOR = {W.J. Fitzgerald and  S. J. Godsill and  A.C. Kokaram and A.J. Stark},
TITLE = {Bayesian methods in signal and image processing (with discussion)},
YEAR = 1999,
GZ = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/98/valencia.ps.gz},
ABSTRACT = {In this paper, an overview of Bayesian methods and models in signal and image
processing is given. The first part of the paper reviews some
traditional classes of model employed for signal processing time
series analysis. Marginal inference based upon analytic integration
of hyperparameters is described for these models and illustrations
are given for the problem of estimating sinusoidal frequency
components in white Gaussian noise and for the general changepoint
problem applied to digital communications. In the second part of the
paper, state of the art applications are described which employ MCMC
methods for the enhancement of noise degraded audio signals,
nonlinear system identification and
image sequence restoration. The complex modelling requirements and large datasets involved in these problems
require sophisticated MCMC schemes employing efficient blocking
schemes, model uncertainty strategies (both reversible jump and
Gibbs variable selection) and nonlinear/non-Gaussian models.}
}


@INCOLLECTION{Clapp_Godsill_1999a,
BOOKTITLE = {Bayesian Statistics VI},
EDITOR = {J.M. Bernardo and J.O. Berger and  A.P. Dawid and A.F.M. Smith},
PUBLISHER = {Oxford University Press},
AUTHOR = {T. C. Clapp and S. J. Godsill},
TITLE = { Fixed-Lag Smoothing using Sequential Importance Sampling},
PAGES = {743--752},
YEAR = 1999,
GZ = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/98/valencia.ps.gz}
}


@INCOLLECTION{Godsill_Clapp_2000,
AUTHOR = {S. J. Godsill and T. C. Clapp},
TITLE = {Improvement strategies for {M}onte {C}arlo particle filters},
BOOKTITLE = {Sequential Monte Carlo Methods in Practice},
EDITOR = {A Doucet and J. F. G. {De Freitas} and N. J. Gordon},
PUBLISHER = {New York: Springer-Verlag},
YEAR = {2001},
PS = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/01/smcbook1.ps}
}


@INPROCEEDINGS{Kokaram_Godsill_1996a,
AUTHOR = {A. C. Kokaram and S. J. Godsill},
TITLE = { A System for reconstruction of missing data in image
sequences using sampled 3{D} {AR} models and {MRF} motion priors.},
BOOKTITLE = {Computer Vision - ECCV '96},
PUBLISHER = {Springer Lecture
Notes in Computer Science},
YEAR = 1996,
MONTH = APR,
PAGES = {613--624},
VOLUME = {II},
GZ = {http://www-sigproc.eng.cam.ac.uk/~ack/intnewa4.ps.gz}
}


@INCOLLECTION{Godsill_2003,
AUTHOR = {S. J. Godsill},
TITLE = {Discussion of Trans-dimensional {M}arkov chain {M}onte
{C}arlo' by {P}eter {J}. {G}reen (in press)},
BOOKTITLE = {Highly Structured Stochastic Systems},
PUBLISHER = {OUP},
YEAR = 2003,
PDF = {http://www.stats.bris.ac.uk/~peter/papers/hssschapter.pdf}
}


@INCOLLECTION{Davy_Godsill_2003,
BOOKTITLE = {Bayesian Statistics VII},
EDITOR = {J.M. Bernardo and J.O. Berger and  A.P. Dawid and A.F.M. Smith},
PUBLISHER = {Oxford University Press},
AUTHOR = {M.Davy and  S. J. Godsill},
TITLE = {Bayesian Harmonic Models for Musical Signal Analysis
(with discussion)},
YEAR = 2003,
PDF = {http://www-sigproc.eng.cam.ac.uk/~sjg/papers/02/harmonicfinal2.ps},
ABSTRACT = {This work is concerned with the Bayesian analysis of musical signals. The
ultimate aim is to use Bayesian hierarchical structures in order to
infer quantities at the highest level, including such things as
musical pitch, dynamics, timbre, instrument identity, etc. Analysis
of real musical signals is complicated by many things, including the
presence of transient sounds, noises and the complex structure of
musical pitches in the frequency domain. The problem is truly
Bayesian in that there is a wealth of (often subjective) prior
knowledge about how musical signals are constructed, which can be
exploited in order to achieve more accurate inference about the
musical structure. Here we propose developments to an earlier
Bayesian model which describes each component note' at a given time
in terms of a fundamental frequency, partials (harmonics'), and
amplitude. This basic model is modified for greater realism to
include non-white residuals, time-varying amplitudes and partials
detuned' from the natural linear relationship. The unknown
parameters of the new model are simulated using a variable dimension
MCMC algorithm, leading to a highly sophisticated analysis tool. We
discuss how the models and algorithms can be applied for feature
extraction, polyphonic music transcription, source separation and
restoration of musical sources.}
}


@BOOK{Godsill_Rayner_1998,
AUTHOR = {S. J. Godsill and P. J. W.  Rayner},
TITLE = {Digital Audio Restoration: A Statistical
Model-Based Approach},
PUBLISHER = {Berlin: Springer, ISBN 3 540 76222 1},
YEAR = {1998},
MONTH = SEP,
ABSTRACT = {The application of digital signal processing (DSP) to problems in audio has been an area of growing importance since the pioneering DSP work of the 1960s and 70s.
In the 1980s, DSP micro-chips became sufficiently powerful to handle
the complex processing operations required for sound restoration in
real-time, or close to real-time. This led to the first commercially
available restoration systems, with companies such as CEDAR Audio
Ltd. in the UK and Sonic Solutions in the US selling dedicated
systems world-wide to recording studios, broadcasting companies,
media archives and film studios. Vast amounts of important audio
material, ranging from  historic recordings of the last century to
relatively recent recordings on analogue or even digital tape media,
were noise-reduced and re-released on CD for the increasingly
quality-conscious music enthusiast. Indeed, the first restorations
were a revelation in that clicks, crackles and hiss could
for the first time be almost completely eliminated from
recordings which might otherwise be un-releasable in CD format.

Until recently, however, digital audio processing has required high-powered computational engines which were only available to large institutions who could afford to use the sophisticated digital remastering technology.
With the advent of compact disc and other digital audio formats, followed by the increased  accessibility of home computing, digital audio processing is now available to anyone who owns a PC with sound card, and will be of increasing importance, in association with digital video, as the multimedia revolution continues into the next millennium. Digital audio restoration will thus find increasing application to sound recordings from the internet, home recordings and speech, and high-quality noise-reducers will become a standard part of any computer system and hifi system, alongside speech recognisers and image processors.

},
URL = {http://www-sigproc.eng.cam.ac.uk/~sjg/springer/index.html}
}


This file has been generated by bibtex2html 1.74