* Remove batchnorm, weight decay, and SOS. * Make --context-size configurable. * Update results.
* Disable weight decay. * Remove input feature batchnorm.. * Replace BatchNorm in the Conformer model with LayerNorm. * Use tanh in the joint network. * Remove sos ID. * Reduce the number of decoder layers from 4 to 2. * Minor fixes. * Fix typos.