SRI International
-
Modeling NERFs for Speaker Recognition
We introduce a new type of feature to capture long-range patterns associated with individual speakers or with speaking styles. NERFs, or Nonuniform Extraction Region Features, are defined based on regions of speech that are delimited by various automatically extractable events of interest. We propose three methods for modeling NERFs that cope with missing features.
-
Visual Odometry
We present a system that estimates the motion of a stereo head or a single moving camera based on video input.
-
Evaluation Of Bias Correction Methods For ‘Worst-Case’ Selective Non-Participation In NAEP
Although NAEP has not been given an official confirmatory role under NCLB, one can expect greater scrutiny of the relationship between state scores and state NAEP scores in the coming years.
-
TRAPping Conversational Speech: Extending TRAP/Tandem Approaches to Conversational Telephone Speech Recognition
In this paper we report experiments with a reduced conversational speech task that led to the adoption of a number of engineering decisions for the design of an acoustic front end. We then describe our results with this front end on a full vocabulary conversational telephone speech task. In both cases the front end yielded…
-
Voicing Feature Integration in SRI’s Decipher LVCSR System
We augment the Mel cepstral (MFCC) feature representation with voicing features from an independent front end.
-
Assigning Adaptive NAEP Booklets Based On State Assessment Scores: A Simulation Study Of The Impact On Standard Errors
This simulation assessed the improvements that could be expected if students were pre-assigned to test booklets that were adapted to their level of performance.
-
Application of the Modified Group Delay Function to Speaker Identification and Discrimination
In this paper, we explore new methods by which speakers can be identified and discriminated, using features derived from the fourier transform phase. A Gaussian mixture model (GMM) based speaker identification system is built with the MODGDF as the front end feature.
-
The Use of a Linguistically Motivated Language Model in Conversational Speech Recognition
In this paper we show that such a model can be used effectively and efficiently in all stages of a complex, multi-pass conversational telephone speech recognition system.
-
Demonstrating Boolean Logic Using Simple Electrical Circuits
To provide a more direct and tangible connection of logic with a real-world application, I developed this enrichment activity.
-
Investigation of Moisture-Assisted Fracture in SiO2 Films Using a Channel Cracking Technique
The technique of channel cracking was developed to study moisture-assisted crack growth in SiO2 films.
-
Cross-dialectal Acoustic Data Sharing for Arabic Speech Recognition
In this paper we describe the use of acoustic data from Modern Standard Arabic (MSA) to improve the recognition of Egyptian Conversational Arabic (ECA).
-
First Experiences with Early Intervention: A National Perspective
This study was designed to determine families’ initial experiences in determining their child’s eligibility, interactions with medical professionals, effort required to obtain services, participation in planning for services, satisfaction with services, and interactions with professionals.