Wednesday, April 24, 2013


WVDP Reports
From August to January…
From August 2012 to January 2013, the West Virginia Dialect Project (WVDP) spent hundreds of hours submerged in Phase Two of their three-phase NSF plan. Phase Two consists of a study of phonetic variation in Appalachia.

Phonetic variation in any region is a daily event. Language is constantly changing to cope with social, geographic, and linguistic pressures. The WVDP is researching the changes in language that have occurred throughout the years, specifically in Appalachia. To get a better social viewpoint on this change, the WVDP is carefully investigating 67 speakers in the West Virginia Corpus of English in Appalachia (WVCEA). These speakers are evenly distributed by age, gender, and geographic location.

During Phase One of the WVDP, each speaker in the WVCEA participated in an interview.  The WVDP team then manually time-aligned each audio file with its transcript, so that the written words flow along with the sounds.  Each time-aligned interview is divided into thousands of utterances; these sound slides can then be analyzed with computer software to assess their acoustic qualities. The time-aligned files are stored on the Sociolinguistic Archive and Analysis Project (SLAAP) database at the NC State Library.


The first step of Phase Two was to search and organize these sound slides from the SLAAP database. The database can be searched based on the orthographic spelling of words spoken within the audio file. Specifically, the team worked with files containing [h]-initial words and [w]-initial words.

Figure 1: A sample view of the SLAAP database
 
What we’re working on now….
Once the proper audio files were identified, the WVDP then began the process of acoustic analysis. Using a acoustic analysis software called Praat, the WVDP primarily focused on the study of two different variations: h-lenition and w/wh-merger.
  
H-lenition is a change of the [h] sound in words like huge and Houston. The [h] was occasionally lost in such words; instead of having a strong, breathy sound, the words might start with the sound. The loss of [h] might also be happening in words like hope and head.

The w/wh-merger is a change where the historically voiceless sound of which is becoming more voiced like the sound of witch. Historically, - and -initial words have had different pronunciations. The [w]-initial sounds in and are examples of how these two sounds are merging.  The WVDP is in the process of analyzing this merger and how this variation has changed over time. Below, the picture shows the w-merger analysis process for one particular speaker in the WVCEA.
 

Figure 2: Praat windows and resulting data



No comments:

Post a Comment