!DOCTYPE html>
9. Clinical Approach to Acoustic Assessment
Keywords
Acoustic assessment in childrenPediatric voice evaluationSpeech-language pathologistVoice recording for assessmentProtocols for instrumental assessment of voiceOverview
Acoustic assessment is an integral part of the pediatric voice evaluation and requires that the speech-language pathologist (SLP) gathers reliable data that can be compared to a normative sample, thereby allowing for objective measurement of dysphonia. It goes without saying, obtaining a reliable data set with children can be challenging. The clinician must be prepared to deviate from standard protocols as needed, and tailor (customize) the assessment tasks to a given child, with consideration for the child’s age, cognitive level, and general level of cooperation. In order to effectively determine appropriate modifications to use during the acoustic assessment, the clinician must have a clear understanding of the principles and foundations of acoustic assessment and analysis. Additionally, assessment tasks should be presented in the order of priority, as attention and cooperation can quickly wane and the clinician needs to be prepared for the child to cease cooperating at any time. Although children can often be coaxed back into finishing an assessment session, some may act silly or stubborn, both of which can result in an affective voice quality. The clinician should note any changes in voice production that may not be related to a voice disorder but rather are attributed to inattention, boredom, or other behavioral factors.
A Note on Age
Can the child imitate?
Is the child willing to interact or engage with the clinician?
Are the parents willing to help?
If the answer is “yes” to any of the above questions, then we recommend trying to collect acoustic data from that child.
Prior to Collecting Voice Recording
The child’s comfort level can have a significant impact on what acoustic recordings are collected, how the child’s voice sounds, and degree of variation from the standard protocol needed. The examination room is unfamiliar and can be an anxiety-provoking place. If a child does not feel comfortable during an assessment, clinicians may hear statements from parents such as “she is much louder at home” or “he is using a ‘baby’ voice right now because he is nervous.” It is important to verify with caregivers that the child’s voice during the assessment is consistent with, or different from, their usual manner of speaking.
An essential part of acquiring a usable acoustic sample is making the child as comfortable as possible. Incorporating toys in the session, engaging the parent, and taking the time to allow the child to acclimate to the clinician and the new environment may be beneficial to a successful outcome. When possible, the clinician should take time at the beginning of the assessment to engage the child in natural conversation, before a microphone is placed in front of him or her. The more comfortable the child is, the more likely we are to obtain a recording of the child’s natural voice and speech patterns.
Equipment and Calibration
Detailed instructions on equipment specifications and setup can be found in the article constructed by an expert panel entitled “Recommended Protocols for Instrumental Assessment of Voice: American Speech-Language-Hearing Association Expert Panel to Develop a Protocol for Instrumental Assessment of Vocal Function” (ASHA-IVAP) [1]. One key element to successful acoustic assessment with children is a head-mounted microphone. This may be a microphone that is affixed to the head via an adjustable headset or one that goes behind the head and over the ears. Using a head-mounted microphone maintains a consistent distance from the microphone to mouth, thereby improving the signal to noise ratio and reducing the overall impact of environmental noise. To minimize noise in the signal further, the child is encouraged to sit as still as possible. It may be beneficial to ask the parents to assist during the assessment, and, for some children, sitting on their parent’s lap may be helpful. When the child tolerates the headset without excessive movement, a calibration value can be acquired prior to completing the tasks, and this allows the clinician to address absolute measures of sound pressure level. See ASHA-IVAP for details on how to perform a calibration [1]. Due to both the added time calibration requires as well as the need for the microphone to stay in a consistent place, calibration can be undertaken with children when there is no concern regarding attentional difficulties or excessive movement. When the calibration is not acquired or the microphone moved during the assessment, absolute measures of sound pressure level cannot be calculated for the assessment.
Some children may not tolerate a head-worn microphone. Whether this is due to sensory difficulties or general compliance, using a microphone attached to a headset is not always feasible. Alternative options include the clinician holding a microphone, placing a microphone on a stand on the table in front of the child, or attaching a microphone to the child’s clothes. If an alternative option for microphone placement is used, the acoustic analyses must be interpreted with caution. These alternative methods may result in inconsistency in the signal due to the child’s variable distance from the microphone.
The clinician should monitor the gain of the signal throughout the assessment, whether the microphone is mounted on the head or in an alternative placement, as many children present with variable loudness. For example, a child may sporadically speak with a loud voice, lean into the microphone to speak, or intermittently speak forcefully to emphasize a point. These aspects may be natural features of a child’s speech; however, they can cause clipping or distortion of the signal. If these elements are intermittent and the gain is appropriate, the clinician can avoid adjusting the gain as long as sections that are clipped are not analyzed. If the signal is consistently clipped, then the gain should be lowered, and that signal should not be used for analysis.
Elicitation of Speech Tasks
It is important for the clinician to consider three points with each task that is undertaken during the assessment: (a) purpose, what is the purpose of the task (i.e., what information will it provide about the child’s voice); (b) optimal elicitation method , what is the standard protocol to collect acoustic data for the task; and (c) alternative elicitation method , if the optimal elicitation method wasn’t feasible, what was the actual elicitation method performed. Answers to these questions provide valuable guidance for the clinician performing the assessment. Understanding the purpose of the task will help the clinician prioritize which tasks to try and elicit, and in what order. When interpreting the results of the completed tasks, it is essential to note if there was any deviation from the standard administration protocol. It is optimal to use a standardized protocol for elicitation and production, thereby allowing comparison of the assessment results to published normative data (e.g., [2, 3]). When the clinician deviates from the standard acquisition protocol, this should be documented, and consideration must be given to this with the acoustic analysis.
Sustained Vowels
Purpose
The speech tasks for analysis include sustained vowel productions elicited at the child’s typical speaking pitch and volume. Measurements that are done include habitual fundamental frequency (fo) and vocal quality measures such as cepstral peak prominence (CPP). Additional measures of vocal quality including jitter, shimmer, and harmonics to noise ratio may be evaluated as well, although they should be in interpreted with caution.
Optimal Elicitation Method
The suggested protocol for sustained vowel assessment is the production of the /ɑ/ vowel for 3–5 s at a steady pitch and volume [1]. Children may require cues to successfully complete the task. These can consist of prompts to take a substantial (big) breath before speaking or may involve the clinician modeling a sustained vowel. When modeling a task, the clinician should be aware that the child may try to exactly imitate the clinician’s voice; i.e., the clinician’s fo may influence the fo produced by the child [4]. The clinician should try to model as infrequently as possible and, when modeling is done, encourage the child to speak at his or her own pitch. Visual cues are also an effective method of assisting children to produce sustained vowels. This may include the clinician silently counting the seconds of the sustained production out on her fingers or slowly bringing her hands together, with contact signaling the cessation of phonation.
One second of the vowel production that is steady in pitch and loudness should be selected for analysis. Research has shown that fo measures in children can be consistently measured from vowels using either time-based measures [5, 6] or frequency-based measures [2, 6]. Time-based measures of vocal quality, including jitter, shimmer, and harmonics to noise ratio, were previously considered to be the standard. These have been found to be less consistent [5, 6] and significantly affected by deviations in loudness [7]. There has been a recent shift to focus on frequency-based measures, specifically cepstral peak prominence (CPP; [1]), which provides an overall measure of periodicity. The measure of CPP has been shown to have good reliability in children [6] and is the current recommended protocol for analysis of sustained vowels [1]. Although measures of CPP are more robust than time-based measures, the clinician must still monitor for steady production of the vowel, as CPP values can be impacted by changes such as loudness [8].
Alternative Elicitation Method
Ghost (“boo”)
Cow (“moo”)
Sheep (“baa”)
Owl (“whoo”)
Train (“choo-choo”)
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
