Trained male singers sang (chest and falsetto registers) and spoke CVCV syllables embedded in a carrier phrase, using as stressed‐unstressed/unstressed‐stressed paradigm. The consonant /t/ was employed throughout in combination with the vowels /i‐a‐u/. All syllables were segmented into three acoustic elements: (1) silent interval, (2) noise burst, and (3) phonation. The silent intervalplusthe noise element were considered to correspond with the consonant, and the phonation segment with the vowel. In appropriate instances, measures were made of duration, amplitude, fundamental frequency, and formant characteristics of the segments from spectrographic records. Relations among segments are reported as a function of modes of utterance and stress. Inferences are drawn concerning possible differential articulatory and phonatory demands on the production system as a consequence of syllabic stress during singing and speech.