Animal and Phoneme Mimic Exercises and Singing Voice Quality

Acoustic and perceptual analysis of how vocal exercises derived from animal and phoneme imitations affect singing voice quality.

Abstract

This study investigates how vocal exercises derived from animal imitations (owl, cat, cow) and phoneme imitations (/u/, /ae/, /a/) affect singing voice quality, measured both acoustically and perceptually.

Study Design

• 5 participants, each recorded in 2 sessions
• Session A — animal mimic protocol: owl, cat, cow imitations → derived exercises → song
• Session B — vowel/phoneme mimic protocol: /u/, /ae/, /a/, /ae/-/u/ → derived exercises → song
• Target song: “Somewhere Over the Rainbow” — recorded pre-session and after each exercise block
• Perceptual evaluation: Omni-VES
• Acoustic measures: f0, jitter, shimmer, HNR, intensity (calibrated), CoG, spectral SD, skewness, kurtosis, CPPS, H1-H2, F1, F2, alpha ratio, L/H ratio, LTAS slope/tilt, AVQI

Analysis Notebooks

Notebook	Description
Acoustic Analysis: Animal vs. Vowel Mimic Exercises	Effect sizes by pair, mimic-to-exercise transfer, rep-by-rep trajectories, carry-over, VES correlations, AVQI

Key Findings (Preliminary)

• Each animal mimic invites a different vocal adjustment: cow changes phonation, cat changes resonance/spectrum, owl changes intensity/stability
• Effect sizes are very large (d > 2 for multiple measures across all three pairs)
• Transfer from mimic to exercise is partial and pair-dependent — cat spectral features transfer best
• Neither protocol harms overall voice quality (AVQI stable or improved)
• Perceived quality (VES) is vowel-dependent and functions more as a task demand scale than a voice quality scale