US11545142B2 - Using Context Information with End-to-End Models for Speech Recognition