Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
2002-02-04
2003-06-24
Banks-Harold, Marsha D. (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C434S185000, C434S322000
Reexamination Certificate
active
06584440
ABSTRACT:
BACKGROUND
A person's ability to hear and understand speech may have a profound impact on their performance in life. Detection and evaluation of speech intelligibility problems can greatly aid in their treatment, thereby increasing a person's capability to understand speakers in their environment. Speech intelligibility tests are essential to detecting and evaluating problems in distinguishing speech from other sounds. Of particular importance are tests designed for children so that early detection of speech intelligibility problems is possible.
There are several aspects to hearing that must be measured in order to properly evaluate a person's ability to distinguish speech. One aspect is the extent to which a person is able to hear what one person is saying in the presence of competing voices or other sounds, the so called “cocktail party problem.” In children, this problem is likely to surface in classroom situations, the very places where children must be able to distinguish the speech of a teacher from the general noise of the classroom, in order to perform well.
For this reason, there is a need for a test to determine the effect(s) of the presence of competing sounds, their types and locations on speech intelligibility in children. Such information is useful in evaluating a base level of speech intelligibility of a child and in analyzing what treatment(s), if any, may improve the child's ability to hear. For example, would a hearing aid or cochlear implant benefit the child more? If a child is listening to a target speaker, is there less of a masking effect of competing sounds when they are removed spatially from the speaker (i.e. spatial release from masking or the increased speech intelligibility observed when the target and competing sounds are spatially separated)? If so, what environmental conditions, such as where the child sits in a classroom, might maximize the child's ability to hear? Is a child's spatial release from masking aided by his or her binaural hearing (i.e., the ability to hear with two ears simultaneously and to extract information from the environment that relies on comparing inputs to the two ears), as with adults?
To enable these evaluations, a speech intelligibility test must be able to measure the extent to which spatial release from masking can occur in a realistic environment such as a classroom, thereby providing a more realistic means of predicting a child's performance in the classroom. Furthermore, because children have notoriously short attention spans, a speech intelligibility test must be able to gain reliable results rapidly while engaging the attention of the child.
Prior studies and tests have focused on adult auditory abilities, and to a limited extent, infant auditory abilities, but little is known regarding the auditory abilities of children. For this reason, the ability of children to process complex acoustic inputs and resolve competition for perception and localization between target sources and interfering competing sounds is not known.
Several studies have suggested that one of the primary differences between adults and children is in their speech intelligibility in complex acoustic situations. But prior tests designed specifically for children, are few and insufficient to the task of (1) engaging children, who generally have short attention spans, to enable a tester to determine what auditory factors (including both type and location of competing sound) are contributing to the child's speech intelligibility level; (2) simulating a child's average noisy classroom environment to enable evaluation under realistic circumstances; (3) providing a clinically reliable test able to link the child's environmental experience to the audiology clinic.
For the foregoing reasons, there is a need for a method and system for clinically testing speech intelligibility in children in the presence of competing sounds that vary by type and location. The test must be fast and engaging and able to provide a reliable measure of a child's speech intelligibility and the extent to which spatial release from masking can occur in a realistic environment, which may include either a quiet room or a noisy room such as a classroom, thereby providing a more realistic means than a simple hearing test of predicting a child's performance in the classroom and information that can be used to tailor a child's treatment.
SUMMARY
The present invention is directed to satisfying this need by providing a method and system for rapid and reliable measurement of a child's speech intelligibility so as to enable measurement of an interference and masking effect of the competing sound by its type and location on an ability of the child to hear the target and to spatially segregate sounds using binaural hearing, that is, the degree to which masking and spatial release from masking can occur given characteristics of the type of competing sound and its location relative to the child and to the target, the measurements informing treatment options to improve the child's speech intelligibility in realistic complex environments.
In one version, the method for testing speech intelligibility of a child comprises providing a set of target words for presenting to the child from an auditory target located in front of the child; selecting the set of target words from a plurality of words by determining which of said plurality of words are within a vocabulary of the child, whereby the set of target words comprises only words familiar to the child; providing a set of picture representations of each of said target words for visually presenting to the child; providing a set of competing sounds of an at least one type of competing sound, including speech, speech-shaped noise, time-reversed speech and modulated speech-shaped noise; specifying a sound level at which a target word is presented so as to ensure hearing by the child in the absence of any of the competing sounds; presenting one of the target words to the child at the sound level; presenting a subset of the picture representations to the child of which one matches the presented target word, the child then responding to the presentation by choosing which picture matches the target word, and recording the child's response; and, repeating the presenting the target word step while furthermore simultaneously presenting the selected type of the competing sound at a location proximate the target and repeating the presenting the picture representations step, over all of the competing sounds selected; and, repeating the presenting the target word step while furthermore simultaneously presenting the selected type of competing sound at a location at a distance away from both the target and the child and repeating the presenting the picture representations step, over all of the competing sounds selected.
In another version, the method further comprises changing the sound level at which the target word is presented adaptively, according to the child's response; changing the sound level comprising determining a direction and an amount of an initial step size by which to change the sound level so as to adapt the change in current sound level to the child's response, the determination being governed by a set of rules.
In another version of the method, presenting the subset of picture representations comprises presenting the pictures on a computer screen at the location of the child; the child choosing which of the subset of pictures matches the target word by pointing a cursor to the chosen picture by moving a mouse and clicking the mouse when the cursor is in position at the chosen picture.
In another version, an apparatus is provided having means for performing one or more of the processes described above.
In another aspect, a program storage device, readable by a machine, tangibly embodying a program of instructions executable by the machine is provided to perform one or more of the processes described above.
In another aspect, an artic
Banks-Harold Marsha D.
King Patricia Smith
Lerner Martin
Wisconsin Alumni Research Foundation
LandOfFree
Method and system for rapid and reliable testing of speech... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for rapid and reliable testing of speech..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for rapid and reliable testing of speech... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3091422