Multiparametric Acoustic Indices AVQI and ABI: Analysis of the Effect of Connected Speech Task Length

This study aims to compare the outcomes of the multiparametric acoustic indices, namely the Acoustic Voice Quality Index (AVQI) and the Acoustic Breathiness Index (ABI), across various lengths of connected speech tasks.

An observational, analytical, cross-sectional study was conducted with a convenience sample of 86 teachers (48 females and 38 males) who had no laryngeal alterations and exhibited neutral or mild vocal deviation (G0 or G1 on the GRBASI scale). Samples of sustained vowel [a] and connected speech through counting from 1 to 20 were collected. These samples were edited into different speech task lengths (ranging from 15 to 42 syllables), and the AVQI and ABI scores were extracted using a script in Praat software. Statistical analysis was performed at a 5% significance level, employing Friedman and Tukey Pairwise tests to compare the different task lengths.

AVQI values demonstrated a progressive increase with the rise in the number of syllables in the speech task. In males, statistically significant differences emerged starting at 33 syllables (counting from 1 to 17). In females, differences were noted beginning at 29 syllables (counting from 1 to 16). For ABI, no statistically significant variations were observed across different speech task lengths in either sex.

The length of the speech task affects AVQI results, indicating the necessity for standardization in vocal assessment to ensure consistency. ABI remained stable and effective for analyzing vocal breathiness in various contexts.