Experiments were conducted to investigate the interdependency of frame rates (30, 15, 10 fps) and audiovisual skew (from +163 to -233 ms1 ). Noised nonsense words like 'abagava' were presented to 20 participants who were asked to identify the middle consonant. At low frame rates (10 fps) consonant perception was impaired when audio ran ahead of video content (skew of -113 to -233ms). When audio lagged video, performance improved monotonically to a maximum at +167ms, where performance equaled 30fps in synch. The results suggest that frame rate and skew are not orthogonal parameters but must both be taken into consideration for AV-delivery. The findings do not support the current notion that 10 fps videos do not adequately capture visual content for speech perception. Participants were able to integrate the given bimodal information as well as the 30 fps condition if the audio channel was subjected to an additional 167ms delay. Author Keywords Audio-Visual Integration, Frame R...