Articulatory Contact Pressure & Vocal Loudness

Description

This data set is derived from 13 adults with typical speech (primary spoken language is American English). The contact pressure data are from a miniature Entran EPI-BO pressure transducer attached to a 0.5 mm thick acrylic palatal mold that is customized for each participant. The transducer was positioned on the alveolar ridge approximately 2 mm posterior to the central incisors and slightly off midline (approximately 2 mm to the left of the midline). For this study, participants were seated in a sound booth with a headset microphone in place to record speech. They produced a set of short (3 syllable) sentences with three experimental phonemes embedded at the start of the second syllable. The phonemes of interest were /t/, /d/, and /s/. These were meaningful English sentences. Participants produced the sentences in four conditions: Habitual loudness, twice as loud as habitual (Loud condition), half as loud as habitual (Soft condition), and Whispered speech. A direct magnitude estimation procedure was used to elicit the loudness variation wherein Habitual was assigned a value of 100, and then participants were instructed to increase or decrease loudness relative to that value. The analysis of these data indicated a fairly consistent change in tongue-to-palate contact pressure when speaking in the Loud condition during which contact pressure increased by at least 30% compared to the Habitual condition. Additionally, contact pressures decreased when produced in the Soft condition, and to a lesser extent, the Whisper condition. These patterns of change in articulatory contact pressure were most notable and consistent for the two stop consonant, /t/ and /d/. The results highlight the interconnected nature of the phonatory and articulatory subsystems for speech. Additionally, the data may provide some insight into why loud-focused speech therapy intervention procedures may provide benefit to individuals with Parkinson's disease or other types of dysarthria. However, confirmation of changes in articulatory contact pressure in these patient populations is needed.

Steps to reproduce

1. Custom palatal mold is made of the mouth and then a 0.5 mm thick acrylic appliance is made using thermoforming coping material 2. Entran EPI-BO pressure transducer is adhered to the acrylic mold using red dental wax. Position of the transducer is approximately 2 mm posterior to the upper central incisors and 2 mm to the right off of the midline (as defined by the space between the central incisors) 3. Pressure signal is amplified, low pass filtered at 50 Hz, and recorded digitally using ADInstruments PowerLab and LabChart software. 4. Acoustic signal is simultaneously recorded and displayed in Labchart. A AKG C410 headset microphone and tube amp were used to acquire this signal. 5. The pressure impulse for the targeted experimental phonemes was selected and the peak pressure identified using LabChart software.

Institutions

Categories

Funding

Licence