基本信息
views: 43
![](https://originalfileserver.aminer.cn/sys/aminer/icon/show-trajectory.png)
Bio
I have adequate experience in building open domain chatbot. I developed several key models for XiaoIce, including pre-training conversation models, neural network-based matching models, and generation models, some of which are published in top tier conferences to facilitate chatbot research.
I am studying applying the Transformer family model for the speech field, including both front-end speech tasks and back-end speech tasks. I published several papers about Transformer based speech recognition and separation papers, some of which have achieved SOTA performance on public benchmarks and make contribution to Microsoft product.
Research Interests
Papers共 16 篇Author StatisticsCo-AuthorSimilar Experts
By YearBy Citation主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSINGno. 99 (2024): 2177-2187
Chinese as a Second Language Research from Different Anglespp.265-292, (2024)
ICML 2023pp.5178-5193, (2023)
CoRR (2023)
Conference of the International Speech Communication Association (2023): 4104-4108
CoRR (2023): 1-8
2022 IEEE Spoken Language Technology Workshop (SLT) (2022): 451-457
Conference of the International Speech Communication Association (2022): 57-61
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 35, NEURIPS 2022 (2022)
ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)pp.7092-7096, (2022)
Load More
Author Statistics
Co-Author
Co-Institution
D-Core
- 合作者
- 学生
- 导师
Data Disclaimer
The page data are from open Internet sources, cooperative publishers and automatic analysis results through AI technology. We do not make any commitments and guarantees for the validity, accuracy, correctness, reliability, completeness and timeliness of the page data. If you have any questions, please contact us by email: report@aminer.cn