Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “SPEECH”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 289 records · Page 16

xEMU Suit Integrated Audio Communications System: Ambient and EVA Pressure Testing System Performance

Testing across several airlock and EVA thermal and pressure scenarios has demonstrated that the Integrated Audio System of NASA’s Exploration Extravehicular Mobility Unit (xEMU) spacesuits transmits and receives intelligible audio communications without the use of a commcap or similar worn device. The xEMU audio system consists of internal loudspeakers and digital microphones (Integrated Communications System –ICS) combined with an adaptive Acoustic Echo Canceller (AEC), outbound voice operated transmission (VOX), and automatic gain control (AGC). Transducers are mounted in an “exploded commcap” configuration with helmet-attached speakers near the ears and three microphones positioned at the collar. The AGC removes inbound audio signals (e.g.,suit, Mission Control, Lander, C&W tones) from the outbound comms stream, reducing echo and feedback (squeal) in low-noise suit environments. The reduction of worn communication equipment increases crewmember comfort, range of movement, and situational awareness. However, test results also highlight the need for proper fan, duct, pump, and gas flow integration with suit acoustics and audio. Ductwork may serve as waveguides for various component and structure-borne noise. Sharply angled ducts can generate turbulent-flow noise. Gas flow from inlets above the crewmember’s head can generate noise when cascading over the faceplate and collar (or commcap) microphones. Sufficient acoustic noise levels (1) require increased gain to boost inbound audio, and (2) may distort signals resulting in AEC disruption or artifacts. Suit-noise levels decline with reduced pressure (density), but then elevated speech and audio effort/power become necessary. Whether the Integrated Audio System, commcap, or other device is used, suit acoustic noise can mask speech in outbound comms. This reduces intelligibility and requires other AECs/devices to suppress comms noise. Yet, adjusting a few components may yield significant improvement. We discuss xEMU audio functionality, demonstrate how acoustical treatment combined with inbound signal conditioning improved clarity during tests, and discuss future modifications.

xEMU↗

xEMU Suit Integrated Audio Communications System: Ambient and EVA Pressure Testing System Performance

Testing across several airlock and EVA thermal and pressure scenarios has demonstrated that the Integrated Audio System of NASA’s Exploration Extravehicular Mobility Unit (xEMU) spacesuits transmits and receives intelligible audio communications without the use of a commcap or similar worn device. The xEMU audio system consists of internal loudspeakers and digital microphones (Integrated Communications System –ICS) combined with an adaptive Acoustic Echo Canceller (AEC), outbound voice operated transmission (VOX), and automatic gain control (AGC). Transducers are mounted in an “exploded commcap” configuration with helmet-attached speakers near the ears and three microphones positioned at the collar. The AGC removes inbound audio signals (e.g.,suit, Mission Control, Lander, C&W tones) from the outbound comms stream, reducing echo and feedback (squeal) in low-noise suit environments. The reduction of worn communication equipment increases crewmember comfort, range of movement, and situational awareness. However, test results also highlight the need for proper fan, duct, pump, and gas flow integration with suit acoustics and audio. Ductwork may serve as waveguides for various component and structure-borne noise. Sharply angled ducts can generate turbulent-flow noise. Gas flow from inlets above the crewmember’s head can generate noise when cascading over the faceplate and collar (or commcap) microphones. Sufficient acoustic noise levels (1) require increased gain to boost inbound audio, and (2) may distort signals resulting in AEC disruption or artifacts. Suit-noise levels decline with reduced pressure (density), but then elevated speech and audio effort/power become necessary. Whether the Integrated Audio System, commcap, or other device is used, suit acoustic noise can mask speech in outbound comms. This reduces intelligibility and requires other AECs/devices to suppress comms noise. Yet, adjusting a few components may yield significant improvement. We discuss xEMU audio functionality, demonstrate how acoustical treatment combined with inbound signal conditioning improved clarity during tests, and discuss future modifications.

xEMU↗

Predicting Team Functioning in Long Term Space Missions Using Acoustic and Linguistic Measures

Maintaining optimal team functioning is critical for long-duration space exploration missions, yet traditional monitoring methods, such as self-reports and wearable sensors, often impose operational burdens or suffer from bias. This paper investigates a non-intrusive speech-based artificial intelligence (AI) framework to predict degradations in team functioning using data from the Human Exploration Research Analog (HERA) of the U.S. National Aeronautics and Space Administration (NASA). Using acoustic features, linguistic descriptors, and semantic embeddings, we evaluate static non-linear and temporal machine learning models to predict both objective (task accuracy) and subjective (self-reported efficacy and cohesion) team functioning outcomes. Results indicate that temporal models outperform static approaches, with prediction of objective task accuracy in Team Interaction Battery (TIB) improving from near chance to 71%. Self-reported outcomes, including team efficacy and cohesion, are predicted more reliably than task performance, achieving balanced accuracies of up to 85.56% and 78.12%, respectively, and are found to be most strongly associated with acoustic features. In a second interdependent task, the MMSEV–EVA, accuracies of up to 78% are achieved using temporal models with acoustic features. Furthermore, incorporating just 1–2 days of team-specific historical data systematically improved performance, and acoustic markers from informal pre-task interactions provided modest predictive gains. Finally, while automated preprocessing yielded viable accuracy, humancorrected data provided moderate performance gains, though transcription error rates did not significantly correlate with model performance. These findings highlight the potential of speech as a passive, high-fidelity monitoring tool for autonomous habitats.

Temporal modeling↗

Fractionated Stereotactic Radiotherapy of Vestibular Schwannomas Accelerates Hearing Loss

Objective: To evaluate long-term tumor control and hearing preservation rates in patients with vestibular schwannoma treated with fractionated stereotactic radiotherapy (FSRT), comparing hearing preservation rates to an untreated control group. The relationship between radiation dose to the cochlea and hearing preservation was also investigated. Methods and Materials: Forty-two patients receiving FSRT between 1997 and 2008 with a minimum follow-up of 2 years were included. All patients received 54 Gy in 27-30 fractions during 5.5-6.0 weeks. Clinical and audiometry data were collected prospectively. From a 'wait-and-scan' group, 409 patients were selected as control subjects, matched by initial audiometric parameters. Radiation dose to the cochlea was measured using the original treatment plan and then related to changes in acoustic parameters. Results: Actuarial 2-, 4-, and 10-year tumor control rates were 100%, 91.5%, and 85.0%, respectively. Twenty-one patients had serviceable hearing before FSRT, 8 of whom (38%) retained serviceable hearing at 2 years after FSRT. No patients retained serviceable hearing after 10 years. At 2 years, hearing preservation rates in the control group were 1.8 times higher compared with the group receiving FSRT (P=.007). Radiation dose to the cochlea was significantly correlated to deterioration of the speech reception threshold (P=.03) but not to discrimination loss. Conclusion: FSRT accelerates the naturally occurring hearing loss in patients with vestibular schwannoma. Our findings, using fractionation of radiotherapy, parallel results using single-dose radiation. The radiation dose to the cochlea is correlated to hearing loss measured as the speech reception threshold.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

A Task-Optimized Neural Network Replicates Human Auditory Behavior, Predicts Brain Responses, and Reveals a Cortical Processing Hierarchy

A core goal of auditory neuroscience is to build quantitative models that predict cortical responses to natural sounds. Reasoning that a complete model of auditory cortex must solve ecologically relevant tasks, we optimized hierarchical neural networks for speech and music recognition. The best-performing network contained separate music and speech pathways following early shared processing, potentially replicating human cortical organization. The network performed both tasks as well as humans and exhibited human-like errors despite not being optimized to do so, suggesting common constraints on network and human performance. The network predicted fMRI voxel responses substantially better than traditional spectrotemporal filter models throughout auditory cortex. It also provided a quantitative signature of cortical representational hierarchy—primary and non-primary responses were best predicted by intermediate and late network layers, respectively. Here the results suggest that task optimization provides a powerful set of tools for modeling sensory systems.

59 BASIC BIOLOGICAL SCIENCES↗

Audiovisual Fusion: Challenges and New Approaches

In this study, we review recent results on audiovisual (AV) fusion. We also discuss some of the challenges and report on approaches to address them. One important issue in AV fusion is how the modalities interact and influence each other. This review will address this question in the context of AV speech processing, and especially speech recognition, where one of the issues is that the modalities both interact but also sometimes appear to desynchronize from each other. An additional issue that sometimes arises is that one of the modalities may be missing at test time, although it is available at training time; for example, it may be possible to collect AV training data while only having access to audio at test time. We will review approaches to address this issue from the area of multiview learning, where the goal is to learn a model or representation for each of the modalities separately while taking advantage of the rich multimodal training data. In addition to multiview learning, we also discuss the recent application of deep learning (DL) toward AV fusion. We finally draw conclusions and offer our assessment of the future in the area of AV fusion.

47 OTHER INSTRUMENTATION↗

Skip-Connected Self-Recurrent Spiking Neural Networks with Joint Intrinsic Parameter and Synaptic Weight Training

As an important class of spiking neural networks (SNNs), recurrent spiking neural networks (RSNNs) possess great computational power and have been widely used for processing sequential data like audio and text. However, most RSNNs suffer from two problems. 1. Due to the lack of architectural guidance, random recurrent connectivity is often adopted, which does not guarantee good performance. 2. Training of RSNNs is in general challenging, bottlenecking achievable model accuracy. To address these problems, we propose a new type of RSNNs called Skip-Connected Self-Recurrent SNNs (ScSr-SNNs). Recurrence in ScSr-SNNs is introduced in a stereotyped manner by adding self-recurrent connections to spiking neurons. The SNNs with self-recurrent connections can realize recurrent behaviors similar to those of more complex RSNNs while the error gradients can be more straightforwardly calculated due to the mostly feedforward nature of the network. The network dynamics is enriched by skip connections between nonadjacent layers. Moreover, we propose a new backpropagation (BP) method called backpropagated intrinsic plasticity (BIP) to further boost the performance of ScSr-SNNs by training intrinsic model parameters. Unlike standard intrinsic plasticity rules that adjust the neuron's intrinsic parameters according to neuronal activity, the proposed BIP method optimizes intrinsic parameters based on the backpropagated error gradient of a well-defined global loss function in addition to synaptic weight training. Here, based on challenging speech, neuromorphic speech, and neuromorphic image datasets, the proposed ScSr-SNNs can boost performance by up to 2.85% compared with other types of RSNNs trained by state-of-the-art BP methods.

97 MATHEMATICS AND COMPUTING↗

Phonetic acquisition in cortical dynamics, a computational approach

Many computational theories have been developed to improve artificial phonetic classification performance from linguistic auditory streams. However, less attention has been given to psycholinguistic data and neurophysiological features recently found in cortical tissue. We focus on a context in which basic linguistic units-such as phonemes-are extracted and robustly classified by humans and other animals from complex acoustic streams in speech data. We are especially motivated by the fact that 8-month-old human infants can accomplish segmentation of words from fluent audio streams based exclusively on the statistical relationships between neighboring speech sounds without any kind of supervision. In this paper, we introduce a biologically inspired and fully unsupervised neurocomputational approach that incorporates key neurophysiological and anatomical cortical properties, including columnar organization, spontaneous micro-columnar formation, adaptation to contextual activations and Sparse Distributed Representations (SDRs) produced by means of partial N-Methyl-D-aspartic acid (NMDA) depolarization. Its feature abstraction capabilities show promising phonetic invariance and generalization attributes. Our model improves the performance of a Support Vector Machine (SVM) classifier for monosyllabic, disyllabic and trisyllabic word classification tasks in the presence of environmental disturbances such as white noise, reverberation, and pitch and voice variations. Furthermore, our approach emphasizes potential self-organizing cortical principles achieving improvement without any kind of optimization guidance which could minimize hypothetical loss functions by means of-for example-backpropagation. Thus, our computational model outperforms multiresolution spectro-temporal auditory feature representations using only the statistical sequential structure immerse in the phonotactic rules of the input stream.

60 APPLIED LIFE SCIENCES↗

2016 AMS Mario J. Molina Symposium

A named symposium to honor Dr. Mario J. Molina was held 10–14 January 2016, as part of the 96th American Meteorological Society (AMS) Annual Meeting in New Orleans, Louisiana. Dr. Molina first demonstrated that industrially produced chlorofluorocarbons (CFCs) decompose in the stratosphere and release chlorine atoms, leading to catalytic ozone destruction. His research in stratospheric chemistry was instrumental to the establishment of the 1987 United Nations Montreal Protocol to ban ozone-depleting substances worldwide. Dr. Molina’s contributions to preserving the planet Earth not only save the atmospheric ozone layer, but also protect the climate by reducing the emissions of greenhouse gases. He was awarded the 1995 Nobel Prize in Chemistry for his pioneering research in understanding the stratospheric ozone loss mechanism. In 2013, President Barack Obama announced Dr. Molina as a recipient of the Presidential Medal of Freedom. The 2016 AMS Molina Symposium honored Dr. Molina’s distinguished contributions to research related to atmospheric chemistry. The symposium contained an integrated theme related to atmospheric chemistry, climate, and policy. Dr. Molina delivered a keynote speech at the Symposium. The conference included invited keynote speeches and invited and contributed oral and poster sessions, and a banquet was held on Tuesday January 12, 2016. The symposium covered all aspects of atmospheric chemistry, with topics including (1) Stratospheric chemistry, (2) Tropospheric chemistry, (3) Aerosol nucleation, growth, and transformation, (4) Aerosol properties, (5) Megacity air pollution, and (6) Atmospheric chemistry laboratory, field, and modeling studies. This DOE project supported 14 scientists, including graduate students, post docs, junior research scientists, and non-tenured assistant professors to attend this symposium.

54 ENVIRONMENTAL SCIENCES↗

Improvement of a Vocal Fold Imaging System

Medical professionals can better serve their patients through continual update of their imaging tools. A wide range of pathologies and disease may afflict human vocal cords or, as they’re also known, vocal folds. These diseases can affect human speech hampering the ability of the patient to communicate. Vocal folds must be opened for breathing and the closed to produce speech. Currently methodologies to image markers of potential pathologies are difficult to use and often fail to detect early signs of disease. These current methodologies rely on a strobe light and slower frame rate camera in an attempt to obtain images as the vocal folds travel over the full extent of their motion.

42 ENGINEERING↗

Orthogonal transform feasibility study

The application of various orthogonal transformations to communication was investigated, with particular emphasis placed on speech and visual signal processing. The fundamentals of the one- and two-dimensional orthogonal transforms and their application to speech and visual signals are treated in detail.

Robinson, G. S.↗

Multipath/RFI/modulation study for DRSS-RFI problem: Voice coding and intelligibility testing for a satellite-based air traffic control system

Analog and digital voice coding techniques for application to an L-band satellite-basedair traffic control (ATC) system for over ocean deployment are examined. In addition to performance, the techniques are compared on the basis of cost, size, weight, power consumption, availability, reliability, and multiplexing features. Candidate systems are chosen on the bases of minimum required RF bandwidth and received carrier-to-noise density ratios. A detailed survey of automated and nonautomated intelligibility testing methods and devices is presented and comparisons given. Subjective evaluation of speech system by preference tests is considered. Conclusion and recommendations are developed regarding the selection of the voice system. Likewise, conclusions and recommendations are developed for the appropriate use of intelligibility tests, speech quality measurements, and preference tests with the framework of the proposed ATC system.

Birch, J. N.↗

Human Factors Research Problems in Electronic Voice Warning System Design

The speech messages issued by voice warning systems must be carefully designed in accordance with general principles of human decision making processes, human speech comprehension, and the conditions in which the warnings can occur. The operator's effectiveness must not be degraded by messages that are either inappropriate or difficult to comprehend. Important experimental variables include message content, linguistic redundancy, signal/noise ratio, interference with concurrent tasks, and listener expectations generated by the pragmatic or real world context in which the messages are presented.

C A Simpson↗

Relative effectiveness of several simulated jet engine noise spectral treatments in reducing annoyance in a TV-viewing situation

An experiment was conducted in order to determine the relative effectiveness of several hypothetical jet engine noise treatments and to test hypothesis that speech interference, at least in part, mediates annoyance in a TV-viewing situation. Twenty-four subjects watched television in a simulated living room. Recorded jet flyover noises were presented in such a way as to create the illusion that aircraft were actually flying overhead. There were 27 stimuli (nine spectra at three overall levels) presented at an average rate of approximately one flight every 2 minutes. Subjects judged the annoyance value of individual stimuli using either a category rating method or magnitude estimation method in each of two 1-hour sessions. The spectral treatments most effective in reducing annoyance were at 1.6 Khz and 800 Hz, in that order. The degree of annoyance reduction resulting from all treatments was affected by the overall sound level of the stimuli, with the greatest reduction at the intermediate overall sound level, about 88 to 89 db(A), peak value. The results are interpreted as supporting the hypothesis that speech interference, at least in part, mediates annoyance with aircraft noise in a TV-viewing situation.

Gunn, W. J.↗

Perception and performance in flight simulators: The contribution of vestibular, visual, and auditory information

The pilot's perception and performance in flight simulators is examined. The areas investigated include: vestibular stimulation, flight management and man cockpit information interfacing, and visual perception in flight simulation. The effects of higher levels of rotary acceleration on response time to constant acceleration, tracking performance, and thresholds for angular acceleration are examined. Areas of flight management examined are cockpit display of traffic information, work load, synthetic speech call outs during the landing phase of flight, perceptual factors in the use of a microwave landing system, automatic speech recognition, automation of aircraft operation, and total simulation of flight training.

Source record↗

Linguistic methodology for the analysis of aviation accidents

A linguistic method for the analysis of small group discourse, was developed and the use of this method on transcripts of commercial air transpot accidents is demonstrated. The method identifies the discourse types that occur and determine their linguistic structure; it identifies significant linguistic variables based upon these structures or other linguistic concepts such as speech act and topic; it tests hypotheses that support significance and reliability of these variables; and it indicates the implications of the validated hypotheses. These implications fall into three categories: (1) to train crews to use more nearly optimal communication patterns; (2) to use linguistic variables as indices for aspects of crew performance such as attention; and (3) to provide guidelines for the design of aviation procedures and equipment, especially those that involve speech.

Goguen, J. A.↗

Voice interactive electronic warning systems (VIEWS) - An applied approach to voice technology in the helicopter cockpit

The cockpit has been one of the most rapidly changing areas of new aircraft design over the past thirty years. In connection with these developments, a pilot can now be considered a decision maker/system manager as well as a vehicle controller. There is, however, a trend towards an information overload in the cockpit, and information processing problems begin to occur for the rotorcraft pilot. One approach to overcome the arising difficulties is based on the utilization of voice technology to improve the information transfer rate in the cockpit with respect to both input and output. Attention is given to the background of speech technology, the application of speech technology within the cockpit, voice interactive electronic warning system (VIEWS) simulation, and methodology. Information subsystems are considered along with a dynamic simulation study, and data collection.

Voorhees, J. W.↗