Conv-Transformer Transducer: Low Latency, Low Frame Rate, Streamable End-to-End Speech Recognition.

Wenyong Huang, Wenchao Hu, Yu Ting Yeung, Xiao Chen

Browse the full Interspeech paper archive.