This study was performed to test the feasibility of an experimental approach for assessing voice changes during exposure to increased +Gz acceleration. Such changes are probably due to mechanical alterations of the structures involved in voice production. This may be relevant to automatic speech analysis for flight control. Because voice control by means of auditory feedback may compensate for acceleration effects, the investigations included conditions with masked auditory feedback. Method: Four male subjects read standard speech test material while seated in a human centrifuge both during sustained + 3 Gz acceleration and in a reference condition. Both test runs were repeated with masking of the auditory feedback by a white noise presented via headphones. Microphone and acceleration signals were recorded on a PC-based dynamic signal acquisition board. Long-time average spectra (LTAS), fundamental frequency (F0), and the frequency of the first, second, and third formant (F1, F2, F3) of the vowels /a/, /o/, and /i/ were extracted from the microphone signal for the different conditions. Results: LTAS clearly differed between the masking conditions, but not between reference and +3 Gz conditions. F0 clearly rose with auditory masking and showed a small increase under +3 Gz acceleration. Several effects of +3 Gz acceleration on formant frequencies were found, all of rather small magnitude. Increased acceleration lowered F2 for the vowel /i/. A decrease of F3 was observed for vowels /i/ and /o/. Conclusion: This pilot study has shown the feasibility of an experimental approach to assessing voice changes during exposure to increased +Gz acceleration. Exposure to +3 Gz showed small effects on F0 and several formant frequencies. A definitive forthcoming study should assess the significance of these effects by investigating a greater number of subjects during exposure to acceleration higher than +3 Gz.