Resource-efficient TDNN Architectures for Audio-visual Speech RecognitionAlexandros KoumparoulisGerasimos Potamianoset al.2021EUSIPCO 2021