Wei Li

An Efficient Streaming Non-Recurrent On-Device End-to-End Model with Improvements to Rare-Word Modeling

Tara N. Sainath, T. Sainath, C. Chiu, 2021, Interspeech.

Using Fluency Representation Learned from Sequential Raw Features for Improving Non-native Fluency Scoring

Wei Li, Xiaohai Tian, Kaiqi Fu, 2022, INTERSPEECH.

A Streaming On-Device End-To-End Model Surpassing Server-Side Conventional Model Quality and Latency

Tara N. Sainath, Z. Chen, Yonghui Wu, 2020, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

Low Latency Speech Recognition Using End-to-End Prefetching

Tara N. Sainath, Wei Li, Bo Li, 2020, INTERSPEECH.

Two-Pass End-to-End Speech Recognition pdf

Tara N. Sainath, Rohit Prabhavalkar, Wei Li, 2019, INTERSPEECH.

Tied & Reduced RNN-T Decoder pdf

Tara N. Sainath, Rami Botros, Wei Li, 2021, Interspeech.