Browse Prior Art Database

Real-world listening effort metric for hearing device fitting

IP.com Disclosure Number: IPCOM000248769D
Publication Date: 2017-Jan-09
Document File: 5 page(s) / 209K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed are a method and system to measure a cochlear implant recipient's level of listening effort using response time (RT) in a speech recognition task, and then provide feedback to the user as well as store the results for the clinician to access. The recipient self-administers the test in real-world situations. The novel approach measures RT in both a closed-set task with a graphical user interface (GUI) for user responses and an open-set verbal task in which the user conversationally responds in voice.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 32% of the total text.

Title

Real-World Listening Effort Metric for Hearing Device Fitting

Abstract Statement

Disclosed are a method and system to measure a cochlear implant recipient’s level of listening effort using response time (RT) in a speech recognition task, and then provide feedback to the user as well as store the results for the clinician to access. The recipient self-administers the test in real-world situations. The novel approach measures RT in both a closed-set task with a graphical user interface (GUI) for user responses and an open-set verbal task in which the user conversationally responds in voice.

Problem Addressed 

Researchers and clinicians are continually seeking ways to improve device processing and fitting techniques for cochlear implants and hearing aids. New developments in hearing devices are most frequently evaluated using speech recognition tests in a controlled laboratory or clinical environment. However, standard speech tests might fail to capture meaningful benefits for at least two reasons: (1) speech recognition is robust and may not be sensitive to differences across device parameters, and (2) laboratory experiments and controlled clinical evaluations do not necessarily translate to real-world improvements.

Known Solutions/Approaches

While improvements in speech recognition can be challenging to demonstrate, another area that has been explored is the potential for hearing devices to reduce listening effort. Some laboratory tests have shown differences in listening effort across conditions using methodologies such as pupillometry (Winn et al., 2015) and dual-task paradigms (Sarampalis, 2009). Although these methods have shown promising results, listening-effort tests that include physiological measurements or dual tasks are not practical in real-world settings.

As an alternative to complex laboratory tests of listening effort, the onset time of a listener’s response may be measured in a standard speech recognition task. Response time (RT) is a simple yet informative metric that captures auditory speed of processing, which is thought to represent at least one domain of listening effort. Several studies have shown differences in RT across different conditions even when speech recognition is stable or near ceiling (Gatehouse & Gordon, 1990; Pals et al., 2015). Furthermore, RT is straightforward to integrate into tests of speech recognition, making it a practical method for assessment of listening effort in real-world situations.

Novelty Statement

The novel contribution is a method and system to measure listening effort using RT in a speech recognition task. The recipient self-administers the test in real-world situations.

Description (Components, Process)

Response Time in a Closed-Set Task with Graphical User Interface (GUI)

To measure RT in a closed-set task, the user of the hearing device privately receives an administered speech recognition test. The audio files for the speech test may be stored on the hearing device processo...