Author: SRI International
-
Effective Use of DCTS for Contextualizing Features for Speaker Recognition
This article proposes a new approach for contextualizing features for speaker recognition through the discrete cosine transform (DCT).
-
Automatic Characterization of Speaking Styles in Educational Videos
We use crowd-sourcing to explore speaking style dimensions in online educational videos, and then propose techniques based solely on acoustic features for automatically identifying a subset of the dimensions.
-
Blended Learning Report
With funding from the Michael & Susan Dell Foundation, SRI’s Center for Technology in Learning studied the adoption of blended learning models in selected schools in California and Louisiana associated with five different charter management organizations during the 2011-12 school year. This research report presents the findings of this formative and summative research effort, including…
-
Computationally-Efficient Endpointing Features for Natural Spoken Interaction with Personal-Assistant Systems
We elicit personal-assistant speech using a recognizer with a dramatically increased endpoint threshold, and find frequent non-final pauses. Based on the new data, we develop low-cost acoustic features to discriminate non-final from final pauses.
-
Global Ethics and Virtual Worlds: Ensuring Functional Integrity in Transnational Research Studies
This paper examines a number of issues in this research context, with particular stress on the challenges posed by transnational experimental projects in virtual worlds and social networks.
-
3D Imaging Reveals Electrodynamics of Polar Cap Aurora
Rishbeth prizewinners Hanna Dahlgren and colleagues investigate the nature of an auroral arc appearing within the deep polar cap region.
-
Nicotine Dependence as a Moderator of Genetic Influences on Smoking Cessation Treatment Outcome
In this secondary analysis of clinical trial data, we examined nicotine dependence severity as a moderator of the effects of 1198 single nucleotide polymorphisms (SNPs) in 53 biologically-relevant gene regions on smoking cessation outcomes.
-
ASR Error Detection Using Recurrent Neural Network Language Model and Complementary ASR
Our goal is to locate errors in an utterance so that the dialogue manager can pose appropriate clarification questions to the users.
-
Coordinated Ionospheric Observations Indicating Coupling between Preonset Flow Bursts and Waves That Lead to Substorm Onset
A critical, long-standing problem in substorm research is identification of the sequence of events leading to substorm expansion phase onset.
-
Tripod Fall: Concept and Experiments of a Novel Approach to Humanoid Robot Fall Damage Reduction
This paper addresses a new control strategy to reduce the damage to a humanoid robot during a fall. Instead of following the traditional approach of finding a favorable configuration with which to fall to the ground, this method attempts to stop the robot from falling all the way to the ground.
-
An Autoencoder with Bilingual Sparse Features for Improved Statistical Machine Translation
We propose a natural autoencoder that maps all the discrete and overlapping sparse features for each SCFG rule into a continuous vector, so that the information encoded in sparse feature vectors becomes a dense vector that may enjoy more samples during training and avoid overfitting.
-
Using the Idea Manager to Promote Coherent Understanding of Inquiry Investigations
This study examines the ideas students generated during an online inquiry investigation and how students used these ideas to construct scientific explanations about everyday phenomena.