Mar, 2022
WeNet 2.0:更高效的端到端语音识别工具包
WeNet 2.0: More Productive End-to-End Speech Recognition Toolkit
Binbin Zhang, Di Wu, Zhendong Peng, Xingchen Song, Zhuoyuan Yao...
TL;DRWeNet 2.0 introduces a unified two-pass framework with bidirectional attention decoders, n-gram based language model, contextual biasing framework, and unified IO to support large-scale data for effective model training, achieving up to 10% relative recognition performance improvement over the original WeNet on various corpora and making available several important production-oriented features.