Publications

Compensation of cepstral features for mismatch due to dissimilar train and test conditions has been critical for good performance in many speech applications. Mismatch is typically due to variability from changes in speaker, channel, gender, and environment. Common methods for compensation include RASTA, mean and variance normalization, VTLN, and feature...

The success of support vector machines (SVMs) for classification problems is often dependent on an appropriate normalization of the input feature space. This is particularly true in topic identification, where the relative contribution of the common but uninformative function words can overpower the contribution of the rare but informative content...

In this paper, we introduce a dialect recognition method that makes use of phonetic models adapted per dialect without phonetically labeled data. We show that this method can be implemented efficiently within an existing PRLM system. We compare the performance of this system with other state-of-the-art dialect recognition methods (both...

This paper presents a series of dialect/accent identification results for three sets of dialects with discriminatively trained Gaussian mixture models and feature compensation using eigen-channel decomposition. The classification tasks evaluated in the paper include: 1)the Chinese language classes, 2) American and Indian accented English and 3) discrimination between three Arabic...

This paper presents a description of the MIT Lincoln Laboratory language recognition system submitted to the NIST 2007 Language Recognition Evaluation. This system consists of a fusion of four core recognizers, two based on tokenization and two based on spectral similarity. Results for NIST?s 14-language detection task are presented for...

This paper describes two experimental protocols for direct comparison on human and machine phonetic discrimination performance in continuous speech. These protocols attempt to isolate phonetic discrimination while controlling for language and segmentation biases. Results of two human experiments are described including comparisons with automatic phonetic recognition baselines. Our experiments suggest...

In this work, we address the question of generating accurate likelihood estimates from multi-frame observations in speaker and language recognition. Using a simple theoretical model, we extend the basic assumption of independent frames to include two refinements: a local correlation model across neighboring frames, and a global uncertainty due to...

Proficiency tests in the forensic sciences are vital in the accreditation and quality assurance process. Most commercially available proficiency testing is available for examiners in the traditional forensic disciplines, such as latent prints, drug analysis, DNA, questioned documents, etc. Each of these disciplines is identification based. There are other forensic...

PVTOL provides an object-oriented C++ API that hides the complexity of multicore architectures within a PGAS programming model, improving programmer productivity. Tasks and conduits enable data flow patterns such as pipelining and round-robining. Hierarchical maps concisely describe how to allocate hierarchical arrays across processor and memory hierarchies and provide a...

Matlab is one of the most commonly used languages for scientific computing with approximately one million users worldwide. Many of the programs written in matlab can benefit from the increased performance offered by multicore processors and parallel computing clusters. The Lincoln pMatlab library (http://www.ll.mit.edu.ezproxyberklee.flo.org/pMatlab) allows high performance parallel programs to...