skip to main content

Attention:

The NSF Public Access Repository (PAR) system and access will be unavailable from 11:00 PM ET on Friday, December 13 until 2:00 AM ET on Saturday, December 14 due to maintenance. We apologize for the inconvenience.


Search for: All records

Award ID contains: 1852316

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. In this study, we investigate how different types of masks affect automatic emotion classification in different channels of audio, visual, and multimodal. We train emotion classification models for each modality with the original data without mask and the re-generated data with mask respectively, and investigate how muffled speech and occluded facial expressions change the prediction of emotions. Moreover, we conduct the contribution analysis to study how muffled speech and occluded face interplay with each other and further investigate the individual contribution of audio, visual, and audio-visual modalities to the prediction of emotion with and without mask. Finally, we investigate the cross-corpus emotion recognition across clear speech and re-generated speech with different types of masks, and discuss the robustness of speech emotion recognition. 
    more » « less
  2. null (Ed.)
  3. Wearable and implantable medical devices ranging from wellness monitors to deep brain stimulators are becoming increasingly vital and ubiquitous. Such devices continuously take measurements, which consumes battery. The power consumption is proportional to the amount of information collected and with the frequency of data collection. High power consumption leads to rapid discharging of battery limiting the usage of these devices. These signals are often transmitted wirelessly for analysis, as well as to keep track of the user’s record, which also significantly increases power consumption. In this project, we evaluated adaptively modifying the rate of data collection on these devices, in other words, the sampling rate, for electrophysiological monitoring as the relevance of the signal changes in time. We carried out these tests using a proof-of-concept prototype developed for this project. In particular, we reviewed the effects of such adaptive sampling on intracellular potentials, and motor unit action potentials (MUAPs). By doing so, we were able to reduce the amount of data by 48.95% and power by 41.50% for the MUAPs with an 8% sample loss within MUAPs, and by 69.20% and 57.14% for intracellular potentials with a 6.75% sample loss. 
    more » « less
  4. Immersion into a virtual environment (VE) goften results in adverse symptoms including nausea, dizziness, and disorientation. These symptoms are an indicator of cybersickness,which is a condition similar to motion sickness experienced in VEs. In this paper, we hypothesized that administered cognitive distraction can accelerate the rate of habituation to a VE. This acceleration, therefore, can lower severity of cybersickness in fewer amount of immersions. To evaluate the impact of cognitive distraction on reducing the effects of cybersickness, we designed a VE and carried out a human subject study with control and experimental groups created through stratified random sampling.Subjects were immersed in our VE on four separate sessions, and our experimental group received cognitive distraction throughout the immersions. Cybersickness was measured using the Simulator Sickness Questionnaire (SSQ) and Presence Questionnaire (PQ). Upon comparing the average SSQ subgroups nausea, oculomotor, and disorientation scores reported by participants for each immersion session, we observed that our experimental group exhibited decrease in cybersickness to a greater extent than that of our control group. We completed t-tests for each of these comparisons, to find that these results are statistically insignificant. We plan to continue with this work by incorporating up to 30 total participants to clarify these findings. 
    more » « less