Description
This longitudinal study aimed to determine whether significant differences existed between the baseline inaccurate signals of the /r/ phoneme for children that eventually acquire or do not acquire /r/. Seventeen participants ages 5-8 who had not acquired /r/ in any of its allophonic contexts were recorded approximately every 3 months from the age of recruitment until they either acquired /r/ in conversation (80% accuracy) or turned eight years old. The recorded audio files were trimmed and labelled using Praat, and signal processing was used to compare initial and final recordings of three allophonic variations of /r/ (vocalic, prevocalic, postvocalic) for each participant. Differences were described using Mel-log Spectral plots. For each age group, initial recordings of participants that eventually acquired /r/ were compared to those of participants that did not acquire /r/. Participants that had not acquired /r/ and had yet to turn eight years old were compared by whether they were perceived to be improving or perceived not to be improving. Significant differences in Mel-log spectral plots will be discussed, and the implications of baseline differences will be highlighted, specifically with respect to the feasibility of identifying predictive markers for acquisition
on-acquisition of the difficult /r/ phoneme.
on-acquisition of the difficult /r/ phoneme.
Download count: 3
Details
Title
- Determining Predictors of Acquisition for /r/ Using Acoustic Signal Processing
Contributors
- Hom, Rachel (Author)
- Weinhold, Juliet (Thesis director)
- Daliri, Ayoub (Committee member)
- College of Health Solutions (Contributor)
- School of International Letters and Cultures (Contributor)
- Barrett, The Honors College (Contributor)
Date Created
The date the item was original created (prior to any relationship with the ASU Digital Repositories.)
2019-05
Resource Type
Collections this item is in