Between-speaker variability and temporal organization of the first formant

Area of Science:

• Acoustic phonetics
• Speech production
• Speaker variability

Background:

• Vocalic intervals exhibit dynamic changes in acoustic properties.
• First formant (F1) trajectories, crucial for vowel perception, can be analyzed for their increasing (positive) and decreasing (negative) dynamics.
• Understanding the factors contributing to between-speaker variability in speech is essential for phonetics and speech technology.

Purpose of the Study:

• To investigate whether positive or negative first formant (F1) dynamics better explain between-speaker variability.
• To compare the explanatory power of F1 dynamics with previously studied intensity dynamics.

Main Methods:

• Vocalic intervals were analyzed for their F1 trajectories.
• F1 trajectories were divided into positive (increasing) and negative (decreasing) dynamics.
• Measures including mean, standard deviation, and sequential variability were calculated for both positive and negative F1 dynamics.
• The proportion of between-speaker variability explained by these measures was assessed.

Main Results:

• Measures of negative F1 dynamics explained a greater proportion of between-speaker variability compared to positive F1 dynamics.
• This finding was highly congruent with a previous study demonstrating similar results for intensity dynamics.
• The results suggest that the way speakers decrease their F1 is a more distinguishing characteristic than how they increase it.

Conclusions:

• Negative F1 dynamics are a significant factor in explaining between-speaker variability in speech.
• Speaker-specific articulation patterns, particularly in the execution of F1 decreases, likely underlie these observed differences.
• The findings contribute to a deeper understanding of the acoustic correlates of speaker identity.