Екі өлшемді ерекшеліктер

Neural end-to-end architectures have been recently proposed for spoken language translation (SLT), following the state-of- the-art results obtained in machine transla- tion (MT) and speech recognition (ASR). Motivated by this …...