LipNet - A Spatio-Temporal Convolution Based Lip Reading Neural Network for Sentence Level Prediction
Changes In Architecture -
- Number of Filters and Layers in Spatio-Temporal Convolution.
- Bidiractional LSTMs instead of GRUs.
- Number of fully connected layers and nodes.
For Model Chheckpoints Mail To: avishekdas539@gmail.com
Link to the Actual Paper : LIPNET: END-TO-END SENTENCE-LEVEL LIPREADING