Article · Wikipedia archive · Last revised Jul 12, 2026

LipNet

LipNet is a deep neural network for audio-visual speech recognition (ASVR). It was created by University of Oxford researchers Yannis Assael, Brendan Shillingford, Shimon Whiteson, and Nando de Freitas. The researchers stated that could match mouth movements to text with 93 percent accuracy, though it was criticized for its test using a limited dataset of words and grammar. It was used in Nvidia's autonomous "backseat driver" prototype Co-Pilot.

Last revised
Jul 12, 2026
Read time
≈ 1 min
Length
124 w
Citations
4
Source
Wikipedia ↗

LipNet is a deep neural network for audio-visual speech recognition (ASVR). It was created by University of Oxford researchers Yannis Assael, Brendan Shillingford, Shimon Whiteson, and Nando de Freitas.1 The researchers stated that could match mouth movements to text with 93 percent accuracy,2 though it was criticized for its test using a limited dataset of words and grammar.3 It was used in Nvidia's autonomous "backseat driver" prototype Co-Pilot.4

References

References

  1. Assael, Yannis M.; Shillingford, Brendan; Whiteson, Shimon; de Freitas, Nando (2016-12-16). "LipNet: End-to-End Sentence-level Lipreading". arXiv:1611.01599 [cs.LG].
  2. "AI that lip-reads 'better than humans'". BBC News. 2016-11-08. Retrieved 2026-05-25.
  3. Vincent, James (November 7, 2016). "Can deep learning help solve lip reading?". The Verge.
  4. Quach, Katyanna. "Revealed: How Nvidia's 'backseat driver' AI learned to read lips". www.theregister.com.