Serialized Output Training for End-to-End Overlapped Speech Recognition.

Naoyuki Kanda, Yashesh Gaur, Xiaofei Wang, Zhong Meng, Takuya Yoshioka

Browse the full Interspeech paper archive.