Visual speech recognition

i.e. machine-assisted lip-reading. One approach to Silent speech interface.

• A recent comprehensive review (also includes coverage of visual speech generation):
  • Sheng, C., Kuang, G., Bai, L., Hou, C., Guo, Y., Xu, X., Pietikäinen, M., & Liu, L. (2022). Deep Learning for Visual Speech Analysis: A Survey (arXiv:2205.10839). arXiv. http://arxiv.org/abs/2205.10839
• SotA (as of 2022-07-05):
  • K R Prajwal, Triantafyllos Afouras, Andrew Zisserman. Sub-word Level Lip Reading With Visual Attention. 2021.
  • word error rate of 22.6%