Sung Vowel Productions in Children with Cochlear Implants and Normal Hearing
Mentor 1
Jing Yang
Start Date
10-5-2022 10:00 AM
Description
The cochlear implant (CI) is a surgically implanted auditory prosthesis device for individuals with severe to profound sensorineural hearing loss. It improves speech and language development in prelingually deafened children. However, due to poorly coded pitch information, CI users still show deficits in perceiving and producing pitch information. The purpose of this study is to examine the vowel acoustics in singing mode in children with CIs with reference to children with normal hearing (NH). Singing samples were collected by research collaborators at Washington University in St. Louis. The CI participants included 15 females and 28 males ages 6.83-11.01. The NH participants included 13 females and 8 males ages 7.01-11.01. They sang a song of their choice (most frequently “Happy Birthday” and “The ABC’s”) to guarantee confidence and optimal singing. All participants were recorded in a silent room without provided imitation or instrument support. The target vowels included English vowels /æ, i, eɪ, (j)u/. The singing samples were first segmented into individual notes. A spectrogram analysis software was used to measure landmark locations of vowel onset and offset for the target vowels. TF32, a time-frequency evaluation software, was used to track the first two formants (F1 and F2) at five equidistant time locations over the course of vowel duration for each target vowel. A formant is an area of acoustic energy representing vocal tract resonance. Vowel normalization was implemented to eliminate the effect of varying vocal tract size on formant frequency values. The formant frequency values will be compared between the children with CIs and NH controls. Speech data was also collected from the participants of this study which, when analyzed, will be used to compare the vowel acoustics in speech mode and singing mode in children with CIs and NH.
Sung Vowel Productions in Children with Cochlear Implants and Normal Hearing
The cochlear implant (CI) is a surgically implanted auditory prosthesis device for individuals with severe to profound sensorineural hearing loss. It improves speech and language development in prelingually deafened children. However, due to poorly coded pitch information, CI users still show deficits in perceiving and producing pitch information. The purpose of this study is to examine the vowel acoustics in singing mode in children with CIs with reference to children with normal hearing (NH). Singing samples were collected by research collaborators at Washington University in St. Louis. The CI participants included 15 females and 28 males ages 6.83-11.01. The NH participants included 13 females and 8 males ages 7.01-11.01. They sang a song of their choice (most frequently “Happy Birthday” and “The ABC’s”) to guarantee confidence and optimal singing. All participants were recorded in a silent room without provided imitation or instrument support. The target vowels included English vowels /æ, i, eɪ, (j)u/. The singing samples were first segmented into individual notes. A spectrogram analysis software was used to measure landmark locations of vowel onset and offset for the target vowels. TF32, a time-frequency evaluation software, was used to track the first two formants (F1 and F2) at five equidistant time locations over the course of vowel duration for each target vowel. A formant is an area of acoustic energy representing vocal tract resonance. Vowel normalization was implemented to eliminate the effect of varying vocal tract size on formant frequency values. The formant frequency values will be compared between the children with CIs and NH controls. Speech data was also collected from the participants of this study which, when analyzed, will be used to compare the vowel acoustics in speech mode and singing mode in children with CIs and NH.
