Speech production and perception across distance in acoustically treated and untreated university lecture halls

Classroom communication depends on talker–listener distance, room acoustics, and background noise, yet these factors are rarely examined jointly in both speech production and perception. This study investigated how distance, room acoustics, and bone-conducted masking affect vocal adaptation and speech intelligibility in two university lecture halls (acoustically treated vs. untreated). Twenty-four adults alternated as talkers and listeners, producing speech at three distances (5, 10, 15 m) under three masking conditions (none, low, high). Voice power level, fundamental frequency, pitch variability, and phonation time ratio were analyzed using linear mixed-effects models, and intelligibility using a binomial generalized mixed model. Voice power increased by 1.76 dB per doubling of distance, and fundamental frequency increased by 91.57 cents relative to each talker’s baseline value, indicating systematic spatial compensation across room and noise conditions, independent of room characteristics. Masking elicited Lombard-like increases in voice power (5.15 dB low; 8.14 dB high), with a room-dependent slope ( ≈ 0.34 dB/dB in the treated room vs. ≈ 0.22 dB/dB in the untreated room). Intelligibility declined with distance (OR = 0.55 per doubling), background noise (OR = 0.40 low; 0.20 high), and in the untreated hall (OR = 0.55). Production adjustments did not fully compensate for perceptual losses, highlighting the importance of acoustic design and noise control in educational spaces.