易江燕

易江燕  /  

  • 职  称: 副高级
  • 电子邮件: jiangyan.yi@nlpr.ia.ac.cn

个人简历

研究方向

语音信息处理、个性化语音生成与鉴别、小数据建模、迁移学习

教育经历

2015.9-2018.6,中国科学院自动化研究所, 模式识别与智能系统, 博士 

2007.9-2010.7,中国社会科学院研究生院,计算语言学,硕士

2003.9-2007.7,云南师范大学,计算机科学与技术,本科

工作经历

2020.10-至今,中国科学院自动化研究所,模式识别国家重点实验室,副研究员

2018.7-2020.10,中国科学院自动化研究所,模式识别国家重点实验室,助理研究员

2011.9-2014.12,阿里巴巴集团,数据科学与技术研究院(iDST),资深算法工程师

教学经历

2021-2022学年()第一学期,《语音信息处理》课程,中国科学院大学人工智能学院,主讲教师 

2018-2019学年()第二学期,《语音交互》课程,中国科学院大学计算机科学与技术学院,主讲教师

获奖情况

语音顶级会议国际会议ICASSP 2021多说话人多风格音色克隆大赛极少样本赛道第一名,2021

第十九届全国信号处理学术年会最佳论文,2019

第十三届全国人机语音通讯学术会议最佳论文,2019

Intel AIDC Beijing Best Poster Award,2018

学术活动        

中国计算机学会 (CCF)语音对话与听觉专委会,委员

全国人机语音通讯学术会议(NCMMSC)常设机构,委员

亚太信号与信息处理协会 语音-语言-音频技术委员会 APSIPA SLA TC member

语音顶级国际会议INTERSPEECH 2020,领域主席

语音顶级国际会议ICASSP 2021,分会主席

语音顶级国际会议INTERSPEECH 2020,分会主席

语音顶级国际会议INTERSPEECH 2019,分会主席

语音重要国际会议APSIPA 2019,出版主席

语音重要国内会议NCMMSC 2019,出版主席

                  

重要国际期刊和会议论文IEEE Transactions on ASLP, IEEE Transactions on SMCS, Speech Communication, IEEE Signal Processing Letters, ICASSP, INTERSPEECH等审稿人

部分代表性期刊论文

[1]     Jiangyan Yi, Jianhua Tao, Zhengqi Wen, Ye Bai: Language-Adversarial Transfer Learning for Low-Resource Speech Recognition. IEEE ACM Trans. Audio Speech Lang. Process. 27(3): 621-630 (2019)

[2]     Jiangyan Yi, Zhengqi Wen, Jianhua Tao, Hao Ni, Bin Liu: CTC Regularized Model Adaptation for Improving LSTM RNN Based Multi-Accent Mandarin Speech Recognition. J. Signal Process. Syst. 90(7): 985-997 (2018)

[3]     Cunhang Fan, Jiangyan Yi*, Jianhua Tao, Zhengkun Tian, Bin Liu, Zhengqi Wen: Gated Recurrent Fusion With Joint Training Framework for Robust End-to-End Speech Recognition. IEEE ACM Trans. Audio Speech Lang. Process. 29: 198-209 (2021)

[4]     Ye Bai, Jiangyan Yi*, Jianhua Tao, Zhengqi Wen, Zhengkun Tian, Shuai Zhang: Integrating Knowledge Into End-to-End Speech Recognition From External Text-Only Data. IEEE ACM Trans. Audio Speech Lang. Process. 29: 1340-1351 (2021)

[5]     Ye Bai, Jiangyan Yi*, Jianhua Tao, Zhengkun Tian, Zhengqi Wen, Shuai Zhang: Fast End-to-End Speech Recognition Via Non-Autoregressive Models and Cross-Modal Knowledge Transferring From BERT. IEEE ACM Trans. Audio Speech Lang. Process. 29: 1897-1911 (2021)

[6]     Cunhang Fan, Jianhua Tao, Bin Liu, Jiangyan Yi, Zhengqi Wen, Xuefei Liu: End-to-End Post-Filter for Speech Separation With Deep Attention Fusion Features. IEEE ACM Trans. Audio Speech Lang. Process. 28: 1303-1314 (2020)

[7]     Yibin Zheng, Jianhua Tao, Zhengqi Wen, Jiangyan Yi: Forward-Backward Decoding Sequence for Regularizing End-to-End TTS. IEEE ACM Trans. Audio Speech Lang. Process. 27(12): 2067-2079 (2019)

[8]     Ye Bai, Jiangyan Yi*, Jianhua Tao, Zhengqi Wen, Cunhang Fan: A Public Chinese Dataset for Language Model Adaptation. J. Signal Process. Syst. 92(8): 839-851 (2020)

[9]     易江燕,陶建华,刘斌,温正棋: 基于迁移学习的噪声鲁棒语音识别声学建模, 《清华大学学报:自然科学版》2018年 第1期。

部分代表性会议论文                                                                                                                                                             

[1]     Jiangyan Yi, Ye Bai, Jianhua Tao, Haoxin Ma, Zhengkun Tian, Chenglong Wang , Tao Wang, Ruibo Fu: Half-Truth: A Partially Fake Audio Detection Dataset. INTERSPEECH 2021

[2]     Jiangyan Yi, Jianhua Tao, Zhengkun Tian, Ye Bai, Cunhang Fan: Focal Loss for Punctuation Prediction. INTERSPEECH 2020: 721-725

[3]     Jiangyan Yi, Jianhua Tao, Ye Bai: Language-invariant Bottleneck Features from Adversarial End-to-end Acoustic Models for Low Resource Speech Recognition. ICASSP 2019: 6071-6075

[4]     Jiangyan Yi, Jianhua Tao: Self-attention Based Model for Punctuation Prediction Using Word and Speech Embeddings. ICASSP 2019: 7270-7274

[5]     Jiangyan Yi, Jianhua Tao, Zhengqi Wen, Ye Bai: Adversarial Multilingual Training for Low-Resource Speech Recognition. ICASSP 2018: 4899-4903

[6]     Jiangyan Yi, Jianhua Tao, Zhengqi Wen, Ya Li: Distilling Knowledge from an Ensemble of Models for Punctuation Prediction. INTERSPEECH 2017: 2779-2783

[7]     Shuai Zhang, Jiangyan Yi, Zhengkun Tian, Ye Bai, Jianhua Tao, Zhengqi Wen: Decoupling Pronunciation and Language for End-to-End Code-Switching Automatic Speech Recognition. ICASSP 2021: 6249-6253

[8]     Zhengkun Tian, Jiangyan Yi, Ye Bai, Jianhua Tao, Shuai Zhang, Zhengqi Wen: FSR: Accelerating the Inference Process of Transducer-Based Models by Applying Fast-Skip Regularization. INTERSPEECH 2021

[9]     Shuai Zhang, Jiangyan Yi, Zhengkun Tian, Ye Bai, Jianhua Tao, Xuefei Liu, Zhengqi Wen: End-to-End Spelling Correction Conditioned on Acoustic Feature for Code-switching Speech Recognition. INTERSPEECH 2021

[10]   Haoxin Ma, Jiangyan Yi, Jianhua Tao, Ye Bai, Zhengkun Tian, Chenglong Wang: Continual Learning for Fake Audio Detection. INTERSPEECH 2021

[11]   Zhengkun Tian, Jiangyan Yi, Ye Bai, Jianhua Tao, Shuai Zhang, Zhengqi Wen: Synchronous Transformers for end-to-end Speech Recognition. ICASSP 2020: 7884-7888

[12]   Ye Bai, Jiangyan Yi, Jianhua Tao, Zhengkun Tian, Zhengqi Wen, Shuai Zhang: Listen Attentively, and Spell Once: Whole Sentence Generation via a Non-Autoregressive Architecture for Low-Latency Speech Recognition. INTERSPEECH 2020: 3381-3385

[13]   Zhengkun Tian, Jiangyan Yi, Jianhua Tao, Ye Bai, Shuai Zhang, Zhengqi Wen: Spike-Triggered Non-Autoregressive Transformer for End-to-End Speech Recognition. INTERSPEECH 2020: 5026-5030

[14]   Ye Bai, Jiangyan Yi, Jianhua Tao, Zhengqi Wen, Zhengkun Tian, Chenghao Zhao, Cunhang Fan: A Time Delay Neural Network with Shared Weight Self-Attention for Small-Footprint Keyword Spotting. INTERSPEECH 2019: 2190-2194

[15]   Ye Bai, Jiangyan Yi, Jianhua Tao, Zhengkun Tian, Zhengqi Wen: Learn Spelling from Teachers: Transferring Knowledge from Language Models to Sequence-to-Sequence Speech Recognition. INTERSPEECH 2019: 3795-3799

[16]   Zhengkun Tian, Jiangyan Yi, Jianhua Tao, Ye Bai, Zhengqi Wen: Self-Attention Transducers for End-to-End Speech Recognition. INTERSPEECH 2019: 4395-4399

[17]   Tao Wang, Ruibo Fu, Jiangyan Yi, Jianhua Tao, Zhengqi Wen, Chunyu Qiang, Shiming Wang: Prosody and Voice Factorization for Few-Shot Speaker Adaptation in the Challenge M2voc 2021. ICASSP 2021: 8603-8607

授权专利        

1.基于口音瓶颈特征的声学模型自适应方法, 发明, 2021,  2 作者, 专利号: ZL 2016 1 1232996.4

2.语音识别中的小数据语音声学建模方法, 发明, 2020,  2 作者, 专利号: ZL 2018 1 0455011.7

3.语音识别中的正则化口音自适应方法, 发明, 2019,  2 作者, 专利号: ZL 2016 1 0971766.3