Relative timing of sound and vision: evaluation and correction

Expectations of the viewing public for the quality of television picture and sound are constantly changing. It is time to attract attention to the relative timing of sound and video components of the television signal as an important point of the viewer 's perception of television programs. It is suggested in the paper to start working on a refinement of recommendation ITU1-R BT.1359-1 which indicates the overall timing tolerance between picture and sound. An approach to monitoring and control of audio-to-video delay in the signal processing module of television receivers or set-top-boxes is suggested in the current paper. This approach is based on mappings between the parameters that determine lip movements of speech sounds, and the parameters computed from processing the audio signal of the same speech sounds. This approach can provide a direct real-time evaluation and in-service correction of audio-to-video delay.