Publications

This paper proposes a framework for spectral enhancement of reverberant speech based on inversion of the modulation transfer function. All-pole modeling of modulation spectra of clean and degraded speech are utilized to derive the linear prediction inverse modulation transfer function (LP-IMTF) solution as a low-order IIR filter in the modulation...

This paper presents a description of the MIT Lincoln Laboratory (MITLL) language recognition system developed for the NIST 2011 Language Recognition Evaluation (LRE). The submitted system consisted of a fusion of four core classifiers, three based on spectral similarity and one based on tokenization. Additional system improvements were achieved following...

This letter proposes a new model for preferential attachment in dynamic directed networks. This model consists of a linear time-invariant system that uses past observations to predict future attachment rates, and an innovation noise process that induces growth on vertices that previously had no attachments. Analyzing a large citation network...

The Multi-modal Early Detection Interactive Classifier (MEDIC) is a triage system designed to enable rapid assessment of mild traumatic brain injury (mTBI) when access to expert diagnosis is limited as in a battlefield setting. MEDIC is based on supervised classification that requires three fundamental components to function correctly; these are...

Autoregressive HMM modeling of spectral features has been proposed as a replacement for standard HMM speech synthesis. The merits of the approach are explored, and methods for enforcing stability of the estimated predictor coefficients are presented. It appears that rather than directly estimating autoregressive HMM parameters, greater synthesis accuracy is...

Anomaly detection in graphs is a relevant problem in numerous applications. When determining whether an observation is anomalous with respect to the model of typical behavior, the notion of "goodness of fit" is important. This notion, however, is not well understood in the context of graph data. In this paper...

Document summarization algorithms are most commonly evaluated according to the intrinsic quality of the summaries they produce. An alternate approach is to examine the extrinsic utility of a summary, measured by the ability of the summary to aid a human in the completion of a specific task. In this paper...

This paper explores both supervised and unsupervised topic modeling for spoken audio documents using only phonetic information. In cases where word-based recognition is unavailable or infeasible, phonetic information can be used to indirectly learn and capture information provided by topically relevant lexical items. In some situations, a lack of transcribed...

Vocal fold vibratory asymmetry is often associated with inefficient sound production through its impact on source spectral tilt. This association is investigated in both a computational voice production model and a group of 47 human subjects. The model provides indirect control over the degree of left-right phase asymmetry within a...

Missing or degraded information continues to be a significant practical challenge facing automatic face representation and recognition. Generally, existing approaches seek either to generatively invert the degradation process or find discriminative representations that are immune to it. Ideally, the solution to this problem exists between these two perspectives. To this...