Automatic Speech Recognition Applications: A Study of Methods for Defining Command Vocabularies

Phoneme-based automatic speech recognition (ASR) technology enables designers to easily create custom command words or phrases that users can employ to request service operations. In this paper, I report results from two experiments concerning important dimensions of these ASR command vocabularies, including command naturalness/appropriateness and command recallability. Ease of recall is a critical dimension for assessing ASR commands used in multi-step applications since service subscribers may be engaged in several different cognitive activities that divide attention. Yet techniques for measuring command recallability can be difficult to implement owing to the time required for data collection and analysis. Results of these studies indicate the the dimension of command “naturalness” and memorability are closely related: under appropriate conditions, the simple procedures associated with measuring command naturalness or appropriateness can predict retrievability of command expressions.

Download Full-text

Speech transformation solutions

Pragmatics & Cognition ◽

10.1075/pc.14.2.16kan ◽

2006 ◽

Vol 14 (2) ◽

pp. 411-442

Author(s):

Dimitri Kanevsky ◽

Sara Basson ◽

Alexander Faisman ◽

Leonid Rachevsky ◽

Alex Zlatsin ◽

...

Keyword(s):

Speech Recognition ◽

Data Collection ◽

General Concept ◽

Significant Progress ◽

Recognition Systems ◽

Time Required

This paper outlines the background development of “intelligent” technologies such as speech recognition. Despite significant progress in the development of these technologies, they still fall short in many areas, and rapid advances in areas such as dictation are actually stalled. In this paper we have proposed semi-automatic solutions — smart integration of human and intelligent efforts. One such technique involves improvement to the speech recognition editing interface, thereby reducing the perception of errors to the viewer. Other techniques that are described in the paper are batch enrollment, which allows the user to reduce the amount of time required for enrollment, and content spotting, which can be used for applications that have repeated content flow, such as movies or museum tours. The paper also suggests a general concept of distributive training of speech recognition systems that is based on data collection across a network.

Download Full-text

Medical Case Rounds: A Medium for Training and Studying Real-Life Decision Making

Proceedings of the Human Factors and Ergonomics Society Annual Meeting ◽

10.1177/154193129503902020 ◽

1995 ◽

Vol 39 (20) ◽

pp. 1330-1333 ◽

Cited By ~ 1

Author(s):

Y. Xiao ◽

P. Milgram ◽

D.J. Doyle

Keyword(s):

Decision Making ◽

Problem Solving ◽

Data Collection ◽

Procedural Knowledge ◽

Real Life ◽

Data Collection And Analysis ◽

Potential Problems ◽

Important Means ◽

Cognitive Activities ◽

Life Decision

Case rounds are a prominent and important means of training in medical domains. As part of a field study of anesthesiologists’ problem solving activities, we audiotaped the discussion of 10 cases spreading over 4 case rounds. This paper describes the method used in data collection and analysis, followed by the major findings of the study. The study identified three types of skills transferred in case rounds: procedural knowledge, sensitivity to precursors of potential problems, and the ability to prepare for contingencies. The study also showed the potential of case rounds in studying cognitive activities in naturalistic settings.

Download Full-text

Acceptable Noise Levels Determined by Traditional and Self-Assessed Methods

Journal of the American Academy of Audiology ◽

10.1055/s-0040-1719092 ◽

2021 ◽

Author(s):

Megan Barnett ◽

Alisha L. Jones ◽

Erin Westbrook

Keyword(s):

Data Collection ◽

Research Design ◽

Repeated Measures ◽

Normal Hearing ◽

The Self ◽

Noise Levels ◽

Data Collection And Analysis ◽

Acceptable Noise Level ◽

Efficient Measurement ◽

Time Required

Abstract Background Many apps have been developed for users to screen their hearing in their own home. The purpose of this study was to investigate the validity and efficiency of a self-assessed acceptable noise level (ANL) in comparison to the traditional ANL measurements. Research Design A within-subject repeated measures research design was utilized. Data Collection and Analysis Sixty-two adults with normal hearing were recruited from Auburn University and the surrounding community. ANLs were measured utilizing the traditional measurement as well as the self-assessed ANL via the Unitron uHear app. Results Within-subject repeated measures of variance revealed no significant differences between traditional ANL measurements and self-assessed ANL measurements. Significant differences were found for time required for testing in each condition, revealing self-assessed testing to be significantly faster. Conclusion The self-assessed ANL measurement via the Unitron uHear app is a valid and efficient measurement of ANL in adults with normal hearing.

Download Full-text

Speech Recognition in Noise in Single-Sided Deaf Cochlear Implant Children on Using Adaptive Digital Microphone Technology

Journal of the American Academy of Audiology ◽

10.1055/s-0040-1718932 ◽

2020 ◽

Author(s):

I. Speck ◽

T. Müller ◽

T. F. Jakob ◽

K. Wiebe ◽

A. Aschendorff ◽

...

Keyword(s):

Speech Recognition ◽

Data Collection ◽

Cochlear Implant ◽

Background Noise ◽

Noise Exposure ◽

Normal Hearing ◽

Speech Stimulus ◽

Data Collection And Analysis ◽

Single Sided Deafness ◽

Speech Recognition In Noise

Abstract Background Previous research demonstrated benefits of adaptive digital microphone technologies (ADMTs) in adults with single-sided deafness (SSD) having a cochlear implant (CI). Children with SSD are especially affected by background noise because of their noise exposure in kindergarten and school. Purpose This article aims to evaluate possible effects of ADMT on speech recognition in background noise in children with SSD who use a CI. Study Sample Ten children between 5 and 11 years of age were included. Data Collection and Analysis Speech recognition in noise was assessed for one frontal distant and two lateral speakers. The speech stimulus was presented at a speech level of 65 dB(A) and noise at a level of 55 dB(A). For the presentation condition with one frontal speaker, four listening conditions were assessed: (1) normal-hearing (NH) ear and CI turned off; (2) NH ear and CI; (3) NH ear and CI with ADMT; and (4) NH ear with ADMT and CI. Listening conditions (2) to (4) were also tested for each lateral speaker. The frontal speaker was positioned directly in front of the participant, whereas the lateral speakers were positioned at angles of 90 degrees and –90 degrees to the participant's head. Results Children with SSD who use a CI significantly benefit from the application of ADMT in speech recognition in noise for frontal distant and for lateral speakers. Speech recognition improved significantly with ADMT at the CI and the NH ears. Conclusion Application of ADMT significantly improves speech recognition in noise in children with SSD who use a CI and can therefore be highly recommended. The decision of whether to apply ADMT at the CI NH ear or bilaterally should be made for each child individually.

Download Full-text