Dynamic Temporal Alignment of Speech to Lips
This addresses the tedious manual task of aligning re-recorded speech with lip movements in postproduction for the film industry, representing a novel method for a known bottleneck.
The paper tackles the problem of automating speech-to-lips alignment in movies by stretching and compressing audio to match lip movements, achieving successful alignment both quantitatively and qualitatively, particularly in cases where original voice is unclear and constant shift methods fail.
Many speech segments in movies are re-recorded in a studio during postproduction, to compensate for poor sound quality as recorded on location. Manual alignment of the newly-recorded speech with the original lip movements is a tedious task. We present an audio-to-video alignment method for automating speech to lips alignment, stretching and compressing the audio signal to match the lip movements. This alignment is based on deep audio-visual features, mapping the lips video and the speech signal to a shared representation. Using this shared representation we compute the lip-sync error between every short speech period and every video frame, followed by the determination of the optimal corresponding frame for each short sound period over the entire video clip. We demonstrate successful alignment both quantitatively, using a human perception-inspired metric, as well as qualitatively. The strongest advantage of our audio-to-video approach is in cases where the original voice in unclear, and where a constant shift of the sound can not give a perfect alignment. In these cases state-of-the-art methods will fail.