Acoustic analyses on German vowels in read speech, recited speech, and singing

On the continuum of hyper- and hypoarticulation, speech production is adjusted based on the environment and the listener to ensure intelligibility with minimal articulatory effort. The current study examines 13 German vowels produced by 16 professional female singers in read speech, recited speech, and singing. Stimuli consisted of ten original extracts from German classical art songs. The vowels' first two formants were analyzed, comparing across the three conditions. Furthermore, Euclidean distances were calculated for all vowels. Results generally indicate more hyperarticulation in recited speech and more centralization in singing. Several exceptions to this trend highlight the necessity for future research.