Campus Map

Jack McLaughlin

Principal Engineer





Research Interests

Machine Learning, Speech Processing, Speech Recognition, Signal Processing


Dr. Jack McLaughlin joined the Applied Physics Laboratory in October of 2001 as a Principal Engineer. He has been involved in a number of programs involving signal classification and feature extraction with applications in areas such as sonar signal analysis, speech classification and cybersecurity. As Principal Investigator, he has led research programs on mine detection and classification (funded by the Office of Naval Research) and programs on speaker identification (funded by Air Force Research Laboratory). Prior to joining APL-UW, Dr. McLaughlin worked at MIT Lincoln Laboratory where he was involved in all aspects of their speaker identification work. This included clustering, channel mismatch studies, classification of short utterances, work with extremely noisy military radio data, and participation in NIST evaluations. Dr. McLaughlin also spent a year working on speech recognition for telephony applications at a startup company before arriving at the Laboratory.


B.S. Physics, Worcester Polytechnic Institute, 1986

M.S. Electrical Engineering, Boston University, 1992

Ph.D. Electrical Engineering, University of Washington, 1997


2000-present and while at APL-UW

A technique for adjusting Gaussian mixture model weights that improves speaker identification performance in the presence of phonemic train/test mismatch

McLaughlin, J., and L. Owsley, "A technique for adjusting Gaussian mixture model weights that improves speaker identification performance in the presence of phonemic train/test mismatch," J. Acoust. Soc. Am., 129, 2423, doi:10.1121/1.3587917, 2011.

More Info

1 Apr 2011

Speaker identification is complicated by cases where training material is phonemically deficient. Misclassifications can result either because subsequent test material from that speaker contains primarily the phonemes missing from the training data or because that test material is phonemically most consistent with another talker's model. This situation can arise in any dialog where, for reasons of brevity and clarity, conventions must be imposed on phraseology. We present here a technique for detecting phonemic deficiencies in a speaker model, and then correcting that model to partially compensate for the biased training data. This technique relies upon a specially constructed universal background model (UBM) from which speaker models are adapted. This UBM is formed by weighting several dozen phoneme GMMs using EM training. As a result, each Gaussian component of the UBM (and of the resulting speaker models) corresponds to a specific phoneme. Analysis of the speaker model weights reveals whether the training data had the typical phonemic variety found in ordinary speech, and if it did not, the weights are adjusted. Using a specially designed corpus created from the TIMIT utterances, we show that this reweighting technique improves performance over non-reweighted models. Results are also given for the Air Traffic Control Corpus.

Application of low-frequency methods for estimating object size

McLaughlin, J., B. Hamschin, and G. Okopal, "Application of low-frequency methods for estimating object size," J. Acoust. Soc. Am., 129, 2663, doi:10.1121/1.3588906, 2011.

More Info

1 Apr 2011

Classification of submerged objects has traditionally been performed using high frequency sonars and imaging techniques. While this permits fine matching of target templates to images acquired in the field, HF methods are necessarily limited in range due to absorption of sound by the water. LF sonars, while offering increased detection range, come with some significant challenges related to the limited bandwidth available. Nonetheless, we show that it is feasible to estimate object size using nonimaging techniques. There are a number of low-frequency phenomena that can be exploited to this end. Among these are edge diffraction in which sharply angled facets of objects ("edges") act like independent, radiating point sources, and helical waves, which can be set up in cylindrical objects. We show that with appropriate postprocessing of these returns, object edges can be localized thus allowing object extent to be assessed. In this paper, we describe our processing system, and then give results when this system is applied to over 40 sequences of returns from a rail system. In each sequence, a single solid, proud cylinder is insonified, and our system reports an estimate of cylinder length and radius. Histograms of these estimates cluster roughly around the true values.

Improving target tracking performance by incorporating classification information

Hanusa, E., W.H. Mortensen, D.W. Krout, and J. McLaughlin,"Improving target tracking performance by incorporating classification information," In Proceedings, MTS/IEEE OCEANS 2010, Seattle, 20-23 September, doi:10.1109/OCEANS.2010.5664500 (MTS/IEEE, 2010).

More Info

20 Sep 2010

This paper presents approaches for incorporating classification information into target tracking algorithms, specifically in a multistatic active sonar context. In addition, this paper describes the framework designed for simulation and classification of return time series from simulated targets and clutter in a realistic underwater environment. The simulated target and clutter returns are integrated into an existing contact-based tracking dataset (TNO Blind dataset) for which time series are unavailable. Simulations compare the integrating classification of contacts at different stages of tracking algorithms. Results show improvements in some tracking metrics with no degradation of the others.

More Publications

Acoustics Air-Sea Interaction & Remote Sensing Center for Environmental & Information Systems Center for Industrial & Medical Ultrasound Electronic & Photonic Systems Ocean Engineering Ocean Physics Polar Science Center