Poster Session #2: UC Ballroom
Measuring vocal jitter: Continuous speech vs. sustained vowels
Presentation Type
Poster
Faculty Mentor’s Full Name
Al Yonovitz
Faculty Mentor’s Department
Communicative Sciences and Disorders
Abstract / Artist's Statement
Jitter is commonly known as the cycle-to-cycle variation in the periodic signal generated at the larynx, more commonly known as the voice box. Jitter can be measured using a number of techniques; however, the most common uses a voice recording that is analyzed by measuring each period of the vocal cycle. Measuring jitter is significant in the practice of Speech-Language Pathology because it can be a determining factor in diagnosing vocal pathologies. When measuring vocal jitter, sustained vowel phonation has been the most common way to obtain values. To obtain sustained vowel phonations, clinicians would have a patient maintain a specific vowel for a sustained length of time. This investigation considers an alternative approach which would have the patient perform continuous speech while being recorded, this, in turn, would allow a more natural depiction of the cycle-to-cycle variation in speech. The voiced phonemes would then be removed and jitter values would be derived. Continuous speech may offer a different perspective to the way vocal jitter is determined because of its more natural depiction of speech. Using a throat microphone, directly on the throat and/or the larynx, will eliminate potential errors. The throat microphone should allow extraction of jitter, from continuous speech, by only extracting voiced phonemes. This research study uses speech output and throat vibrational output (throat microphone and standard microphone, respectively) to determine jitter values for sustained vowels and continuous speech. Eight subjects (ages 8-30) produced nine sustained vowel sounds and, additionally, read a standard passage. The data were recorded with a sample rate of 44,100 samples per sec (resolution of 16 bits). Jitter values were obtained for sustained vowels and continuous speech. The results will be discussed for both methods using jitter values in normal speakers.
Measuring vocal jitter: Continuous speech vs. sustained vowels
UC Ballroom
Jitter is commonly known as the cycle-to-cycle variation in the periodic signal generated at the larynx, more commonly known as the voice box. Jitter can be measured using a number of techniques; however, the most common uses a voice recording that is analyzed by measuring each period of the vocal cycle. Measuring jitter is significant in the practice of Speech-Language Pathology because it can be a determining factor in diagnosing vocal pathologies. When measuring vocal jitter, sustained vowel phonation has been the most common way to obtain values. To obtain sustained vowel phonations, clinicians would have a patient maintain a specific vowel for a sustained length of time. This investigation considers an alternative approach which would have the patient perform continuous speech while being recorded, this, in turn, would allow a more natural depiction of the cycle-to-cycle variation in speech. The voiced phonemes would then be removed and jitter values would be derived. Continuous speech may offer a different perspective to the way vocal jitter is determined because of its more natural depiction of speech. Using a throat microphone, directly on the throat and/or the larynx, will eliminate potential errors. The throat microphone should allow extraction of jitter, from continuous speech, by only extracting voiced phonemes. This research study uses speech output and throat vibrational output (throat microphone and standard microphone, respectively) to determine jitter values for sustained vowels and continuous speech. Eight subjects (ages 8-30) produced nine sustained vowel sounds and, additionally, read a standard passage. The data were recorded with a sample rate of 44,100 samples per sec (resolution of 16 bits). Jitter values were obtained for sustained vowels and continuous speech. The results will be discussed for both methods using jitter values in normal speakers.