Charlie Farrington

Durational and spectral differences in American English vowels: Dialect variation within and across regions

Spectral differences among varieties of American English have been widely studied, typically recognizing three major regionally diagnostic vowel shift patterns [Labov, Ash, and Boberg (2006). The Atlas of North American English: Phonetics, Phonology and Sound Change (De Gruyter, Berlin)]. Durational variability across dialects, on the other hand, has received relatively little attention. This paper investigates to what extent regional differences in vowel duration are linked with spectral changes taking place in the Northern, Western, and Southern regions of the U.S. Using F1/F2 and duration measures, the durational correlates of the low back vowel merger, characteristic of Western dialects, and the acoustic reversals of the front tense/lax vowels, characteristic of Southern dialects, are investigated. Results point to a positive correlation between spectral overlap and vowel duration for Northern and Western speakers, suggesting that both F1/F2 measures and durational measures are used for disambiguation of vowel quality. The findings also indicate that, regardless of region, a durational distinction maintains the contrast between the low back vowel classes, particularly in cases of spectral merger. Surprisingly, Southerners show a negative correlation for the vowel shifts most defining of contemporary Southern speech, suggesting that neither spectral position nor durational measures are the most relevant cues for vowel quality in the South.

Evaluating acoustic speaker normalization algorithms: Evidence from longitudinal child data

Speaker vowel formant normalization, a technique that controls for variation introduced by physical differences between speakers, is necessary in variationist studies to compare speakers of different ages, genders, and physiological makeup in order to understand non-physiological variation patterns within populations. Many algorithms have been established to reduce variation introduced into vocalic data from physiological sources. The lack of real-time studies tracking the effectiveness of these normalization algorithms from childhood through adolescence inhibits exploration of child participation in vowel shifts. This analysis compares normalization techniques applied to data collected from ten African American children across five time points. Linear regressions compare the reduction in variation attributable to age and gender for each speaker for the vowels BEET, BAT, BOT, BUT, and BOAR. A normalization technique is successful if it maintains variation attributable to a reference sociolinguistic variable, while reducing variation attributable to age. Results indicate that normalization techniques which rely on both a measure of central tendency and range of the vowel space perform best at reducing variation attributable to age, although some variation attributable to age persists after normalization for some sections of the vowel space.

Contextualizing the Corpus of Regional African American Language:DC

Washington DC has been home to a majority African American population since the late 1950s, with many in the United States considering DC an African American cultural center. At the same time, in the 1960s, it became a key site for foundational studies of African American Language (AAL), spearheaded by researchers from the Center for Applied Linguistics and Georgetown University, as the field of sociolinguistics was in its infancy. This paper is composed of two main parts: first, it provides relevant socio-cultural and demographic information on the history of African Americans in DC; second, this paper provides an overview of the history of linguistic research on AAL in DC and surrounding areas. Research on DC AAL has proceeded steadily if somewhat sporadically since the 1950s, with some studies achieving wider circulation than others. This article thus provides relevant linguistic and socio-historical contextual information for readers interested in learning about DC AAL in connection with the introduction of the Corpus of Regional African American Language, whose core component is centered on DC AAL.

Vowel Dynamics in the Southern Vowel Shift

Southern varieties of English are known to be affected by the Southern Vowel Shift (SVS), which alters the positional relationship between the front tense/ lax system. However, previous work on the SVS generally limits its focus to steady state formant measures. Possible links between these shifts and dynamic trajectory distinctions have largely been unexplored despite widespread recognition that Southern vowels are dynamic in nature. The current article uses data from three Southern states (Tennessee, North Carolina, and Virginia) to ask to what extent does spectral onset position (the typical measure of SVS participation) correlate with internal spectral dynamics in the SVS. Analysis methods include a series of spectral measures (vector length, trajectory length, spectral rate of change and vector angle), which capture vowel inherent dynamics and vowel directionality. Results support the utility of looking at dynamic measures to better understand the fuller extent of vowel changes that occur with the SVS and lend support to recent calls to include nonstatic measures in sociophonetic analyses more generally. keywords: sociophonetics, spectral dynamics, Southern drawl, vowel inherent spectral change, regional variation Speech is not a static process, but an active one, and it is clear that many properties cannot be understood unless we examine their dynamic aspects. —Ladefoged and Maddieson (1996, 6) The southern vowel shift (SVS), a vocalic pattern characterizing English in the southeastern United States, has been well documented since Labov, Yaeger, and Steiner’s (1972) early acoustic work (e.g., Feagin 1986; Thomas 1989, 2001; Labov 1991, 1994, 2001; Fridland 1999, 2001, 2012; Dodsworth and Kohn 2012; Fridland and Kendall 2012, 2015; Kendall and Fridland 2012; Koops 2014). In brief, the SVS refers to a number of changes that affect several parts of the vowel system, including the acoustic repositioning of the front lax and tense vowel pairs /i / ~ /I/ and /e/ ~ /E/, high and mid Downloaded from https://read.dukeupress.edu/american-speech/article-pdf/93/2/186/534375/0930186.pdf by guest on 08 April 2019 Vowel Dynamics in the Southern Vowel Shift 187 back vowel fronting, and /aI/ monophthongization. The changes in the front vowel subsystem are particularly relevant in terms of distinguishing the contemporary Southern vowel system and will be the focus of the current work. The majority of sociophonetic research on vowels has focused on the first two formants (F1 and F2) and, in particular, static measurements of these formants at vowel nuclei. These are known to reflect primary cues to vowel identity (Labov, Yaeger, and Steiner 1972; Thomas 2001) and indeed represent key aspects of vowel quality. However, as the introductory Ladefoged and Maddieson (1996) quotation suggests, speech is an active process and its dynamic aspects need to be examined if we are to understand fully its properties. As such, recent sociophonetic work has considered a wider range of acoustic features in characterizing regional (and other) vowel differences (Fox and Jacewicz 2009; Wassink 2006, 2015). In this vein, the current project seeks to utilize measures that capture different aspects of vowel differences to better understand the role that dynamics play in the SVS. In folk linguistics, Southern speakers are often described as speaking with a “drawl,” characterized by slow speech and “drawn out vowels” (Preston 1986, 1989, 1993), suggesting dynamics are socially salient cues for Southern speech. While little sociolinguistic research has empirically examined the “Southern drawl” (Allbritten 2011; Koops 2014), this kind of folk commentary likely draws, at least somewhat, from an acoustic reality. In fact, the front lax vowels /I/ and /E/, two of the vowels often undergoing SVS changes, are found to have longer durations in the South when compared to other regions (Clopper, Pisoni, and de Jong 2005; Fox and Jacewicz 2009; Fridland, Kendall, and Farrington 2014), as well as exhibiting “breaking,” or becoming triphthongal (Feagin 1986; Allbritten 2011; Koops 2014). However, beyond sporadic mention of the Southern drawl (Sledd 1966; Feagin 1986, 1987; Wetzell 2000; Allbritten 2011) or studies of /aI/ monophthongization (Fridland 2003; Thomas 2003), very little linguistic work on Southern speech has focused on dynamics. The traditional characterization of the SVS is based on single point F1/F2 measures, usually in terms of a vowel nucleus or midpoint, despite wide acknowledgment that vowels in Southern speech are characterized by different dynamics than vowels elsewhere (Fridland 2012). It is surprising that there is a distinct lack of work done on vowel dynamic properties despite the potential relevance of trajectory information in maintaining phonemic distinctions for the front tense and lax vowel pairs (Kingston and Diehl 1994). This leads to the present project, which focuses on a primary research question: to what extent is the SVS characterized by differences that involve dynamics and not just vowel positions? The current analysis addresses this question about the role of dynamics in the SVS by applying a suite of quantitaDownloaded from https://read.dukeupress.edu/american-speech/article-pdf/93/2/186/534375/0930186.pdf by guest on 08 April 2019 american speech 93.2 (2018) 188 tive measures to production recordings from Southern speakers from three states with a focus on the front vowel subsystem. These measures capture (1) spectral onset position, (2) vowel inherent dynamics, and (3) glide directionality to examine how the SVS is characterized by vowel dynamics beyond typical nuclei measures. Results indicate that while single point measures can identify speakers as shifted or nonshifted, a number of other, nonstatic features also correlate with those measures, likely providing a more robust set of cues to Southern shifted speech. The quantitative measures that reflect dynamicity and glide direction, used together, provide a more complete encapsulation of what it means when we say a speaker has Southern shifted (SVS) vowels and better capture what specific linguistic cues are conveying the folk linguistic sense of “the drawl.” While our substantive focus here is on the dynamic properties of the SVS affected vowels, this work offers more evidence of the import of looking beyond F1/F2 measures for sociophonetic studies of vowels for other language varieties.