Lance W. Porter, Department of Speech and Language Pathology
In a world where more and more opportunities are opening up on an international level, it has become increasingly important to be able to communicate effectively. Accent reduction therapy was developed to help people reduce foreign or regional accents that would otherwise put them at a disadvantage. International students who are speakers of English as a second language (ESL), upon starting school in the United States, will often struggle because of strong accented speech in the already competitive world of academics. Strong accents can adversely effect their ability to communicate and succeed in the classroom. Upon graduation, the ability to be understood clearly is even more important as they look for a job. Sadly enough, accent and ethnicity have been used as negative predictors of employability. However, research (Hamers & Blanc, 1989) has shown that when a well developed standard English accent is used by a speaker of English as a second language the overall perception is highly favorable. Spectrographs have been successfully used to identify acoustic cues important in the description and evaluation of many communication disorders (Eaken & Daniloff, 1991). The increased integration of acoustic cue analysis into accent reduction therapy seems to be a natural progression in the advancement of this field. By identifying spectrographic acoustic cues associated with many of the mistakes made by foreign speakers of English as a second language, the clinician would be provided with a powerful tool to make therapy more effective. Assessment could become more efficient by providing the clinician with actual visualification of clients misarticulations. This information could then be used to specify the focus and direction of a therapy program. In therapy, ESL clients would be provided with a visual comparison of their utterances compared to a native American speaker. This comparison would provide an example that the client could work towards and therefore stimulate the desire to succeed. Finally, spectrographic acoustic cues could be used as baseline and post-treatment data to evaluate the progress of the client and measure effectiveness of therapy.
A subject pool of native French speakers was identified using filers that were placed in the JKHB located at Brigham Young University. Out of the nine people that responded, two were chosen to participate. A short screening instrument, verified that the subjects had a good command of the English language, but had strong accents and demonstrated several of the acoustic cues targeted in this study. Baseline high fidelity speech samples were then recorded inside a sound treated chamber in the BYU Hearing and Speech Sciences Laboratory using a Panasonic digital audio tape recorder. These pre-treatment samples were then acoustically analyzed using the Kay Elemetrics Sona-Graph. Voice onset time in milliseconds, stop-gap in milliseconds, and the relative intensity, timing and priodicity of frication all were examined. The subjects were then involved in traditional accent reduction speech therapy two times a week where imitative interactive and independent massed practice drills were the core experimental treatment. After a period of one week post-treatment, high fidelity speech samples were recorded and analyzed to determine the progress of each subject.
This study indicated that the integration of spectrographic acoustic cue analyses in foreign accent reduction therapy could provide important information during three phases of therapy. First, through the analyses of the pre-treatment language sample I was provided with a visual identification of the subjects’ mistakes and how frequently they occured. This allowed me to choose therapy targets that, when treated, would make the biggest impact on the subjects’ intelligibility. This method of individualized diagnosis proved to be especially important as there were several acoustic cues targeted in this study, but only a limited amount of time available for treatment. This could also prove advantageous in the work field where a client’s time and finances will often limit how much time they can spend in therapy. Next, spectrographic acoustic cue analyses proved useful during the actual accent reduction therapy. By teaching the subjects themselves to identify and monitor the targeted acoustic cues and comparing them to productions made by a native speaker of English, they were provided with accurate feedback and powerful reinforcement. This motivated that subjects to participate in the therapy program. It was discovered, however, that a correct and consistent clinician model was essential for this outcome. During therapy it was also disovered that acoustic cue analyses could provide clues that would help determine how the client was misusing the oral mechanism and what the clinician could do to provide a more effective model. Finally, spectrographic acoustic cue analyses was used to monitor progress by comparing pre-treatment recordings to post-treatment recordings. During this study, both subjects made improvement in many of the targeted acoustic cues. Time did not permit judges to independently rate each speaker’s final productions, but would be important in future studies to establish inter-rater reliability. In addition, on those cues that the subjects continued to struggle with, the post-treatment comparisons could help the clinician determine future therapy targets and methods.
This project is a result of a desire to become more involved and gain a deeper understanding of speech-language pathology. Through a grant from the Office of Research and Creative Work I was able to pursue an interest in the French language and in the application of speech science to improve clinical processes off case management. As a pilot study it will form the base for future work in accent reduction. Under the instruction and guidance of Dr. Laurence M. Hilton I have gained invaluable experience that will help me as I pursue both the Master’s and Doctoral degrees in Speech-Language Pathology.