SRI International
-
Segmentation of Aerial Surveillance Video Using a Mixture of Experts
We propose a video segmentation algorithm for aerial surveillance video. The algorithm uses a Mixture of Experts consisting of a supervised image segmentation algorithm named the Trainable Sequential MAP segmentation algorithm.
-

Advanced modular incoherent scatter radar
Our AMISR ushered in relocatable radar to read atmospheric events, rain, sleet or tsunami.
-
A Cognitive Framework for Delegation to an Assistive User Agent
We present a BDI-based framework for a cognitive agent that acts as an assistant to a human user by performing tasks on her behalf. This framework accommodates: (1) user delegation of tasks to the agent, which may be inconsistent or infeasible or both, and (2) user-specified guidance and preferences on the execution and management of…
-
Incorporating Tandem / HATs MLP Features into SRI’s Conversational Speech Recognition System
We describe the development of a speech recognition system for conversational telephone speech (CTS) that incorporates acoustic features estimated by multilayer perceptrons (MLPs). The acoustic features are based on frame-level phone posterior probabilities, obtained by merging two different MLP estimators, one based on PLP-Tandem features, the other based on hidden activation TRAPs (HATs) features.
-
Constructing Padi Measurement Models for the BEAR Scoring Engine (Padi Technical Report 7)
Connections among the task design, student work products, evaluation, and inferential reasoning are highlighted, and the BEAR Scoring Engine is presented as one example of how multidimensional item response modeling can be integrated with the PADI design system
-
A* Based Joint Segmentation and Classification of Dialog Acts in Multiparty Meetings
We investigate the use of the A* algorithm for joint segmentation and classification of dialog acts (DAs) of the ICSI Meeting Corpus. The proposed method is evaluated on both traditional and new metrics, and compared with our previous work on the same task.
-
Combining Feature Sets with Support Vector Machines: Application to Speaker Recognition
In this paper, we describe a general technique for optimizing the relative weights of feature sets in a support vector machine (SVM) and show how it can be applied to the field of speaker recognition. Our training procedure uses an objective function that maps the relative weights of the feature sets directly to a classification…
-
Four Weightings and a Fusion: A Cepstral-SVM System for Speaker Recognition
A new speaker recognition system is described that uses Mel-frequency cepstral features. This system is a combination of four support vector machines (SVMs). All the SVM systems use polynomial features and they are trained and tested independently using a linear inner-product kernel.
-
Multirate ASR Models for Phone-Class Dependent N-Best List Rescoring
In this work, we describe a technique to augment a recognizer that uses this compromise with information from multiple-rate spectral models that emphasize either better time or better frequency resolution in order to improve performance.
-
VERL: An ontology framework for representing and annotating video events
This article describes the findings of a recent workshop series that has produced an ontology framework for representing video events-called Video Event Representation Language (VERL) -and a companion annotation framework, called Video Event Markup Language (VEML). One of the key concepts in this work is the modeling of events as composable, whereby complex events are…
-
Task Management under Change and Uncertainty: Constraint Solving Experience with the CALO Project
We outline the challenges and opportunities presented by constraint solving in the presence of change and uncertainty, embodied in CALO’s personalized time management and task reasoning and execution systems.
-
Singapore Tablet PC Program Study: Executive Summary and Final Report, Volume 1, Technical Findings