Picture for Zhengqi Wen

Zhengqi Wen

FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization

Apr 07, 2021
Zhengkun Tian, Jiangyan Yi, Ye Bai, Jianhua Tao, Shuai Zhang, Zhengqi Wen

Figure 1 for FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization

Figure 2 for FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization

Figure 3 for FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization

Figure 4 for FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization

* Submitted to INTERSPEECH2021

Access Paper or Ask Questions

TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

Apr 04, 2021
Zhengkun Tian, Jiangyan Yi, Jianhua Tao, Ye Bai, Shuai Zhang, Zhengqi Wen, Xuefei Liu

Figure 1 for TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

Figure 2 for TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

Figure 3 for TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

Figure 4 for TSNAT: Two-Step Non-Autoregressvie Transformer Models for Speech Recognition

* Submitted to Interspeech2021

Access Paper or Ask Questions

Deep Time Delay Neural Network for Speech Enhancement with Full Data Learning

Nov 11, 2020
Cunhang Fan, Bin Liu, Jianhua Tao, Jiangyan Yi, Zhengqi Wen, Leichao Song

Figure 1 for Deep Time Delay Neural Network for Speech Enhancement with Full Data Learning

Figure 2 for Deep Time Delay Neural Network for Speech Enhancement with Full Data Learning

Figure 3 for Deep Time Delay Neural Network for Speech Enhancement with Full Data Learning

Figure 4 for Deep Time Delay Neural Network for Speech Enhancement with Full Data Learning

* Accepted by ISCSLP 2021

Access Paper or Ask Questions

Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition

Nov 09, 2020
Cunhang Fan, Jiangyan Yi, Jianhua Tao, Zhengkun Tian, Bin Liu, Zhengqi Wen

Figure 1 for Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition

Figure 2 for Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition

Figure 3 for Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition

Figure 4 for Gated Recurrent Fusion with Joint Training Framework for Robust End-to-End Speech Recognition

* Accepted by IEEE/ACM Transactions on Audio, Speech, and Language Processing

Access Paper or Ask Questions

Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

May 30, 2020
Ye Bai, Jiangyan Yi, Jianhua Tao, Zhengkun Tian, Zhengqi Wen, Shuai Zhang

Figure 1 for Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

Figure 2 for Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

Figure 3 for Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

Figure 4 for Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition

* submitted to INTERSPEECH2020

Access Paper or Ask Questions