In sociophonetic research there is a strong focus on the acoustic analysis of vowels. For the visualization and normalization of variation in vowels, we developed Visible Vowels (Heeringa & Van de Velde 2018), which is available as a web applicationi and an R packageii. It can be used for a wide range of linguistic sub-disciplines (phonetics, sociolinguistics, dialectology, language acquisition, psycholinguistics). The sociophonetic bias towards the analysis of vowels is partly caused by the lack of phonetic tools for the acoustic analysis of consonants. With the development of Visible Consonants, we want to fill part of this gap and provide the linguistic research community with a user-friendly and free tool for the visualization and normalization of variation in consonants.

In this poster we present a first (beta) version of Visible Consonants.

In Visible Consonants two groups of acoustic measures are distinguished. The first group includes one value acoustic measurements, i.e. there is one unique measurement per consonant such as duration, the energy of a sound in airiii, the power of a sound in airiv, zero crossings, band energy, band density, band energy difference, band density difference, centre of gravity in the spectrum, standard deviation of the spectrum, skewness of the spectrum, kurtosis of the spectrum and central moment.

The second group includes dynamic temporal measurements which are measured at different time points within the consonant, such as amplitude, energy, power, F0, intensity, formants, LPC/LFCC (cepstral coefficients distributed over the interval), MFCC (Mel-frequency cepstral coefficients distributed over the interval). Several dispersion measures can be applied to the dynamic measurements such as means, medians, root mean squares, standard deviations, minimums, maximums, etc.

We aimed for a flexible data format which allows to include an unlimited number of one value variables, dynamic temporal variables and sociolinguistic variables.

The first version of Visible Consonants is limited to acoustic data. In a second phase we also want to be able to process articulatory and aerodynamic data.

We want to create a dialogue with the audience and hope to receive suggestions so that we can meet the specific needs of scholars in variationist linguistics.