ERROR ANALYSIS APPLIED TO END-TO-END SPOKEN LANGUAGE UNDERSTANDING

被引：0

作者：

Caubriere, Antoine ^{[1
]}

Ghannay, Sahar ^{[2
]}

Tomashenko, Natalia ^{[3
]}

De Mori, Renato ^{[3
,4
]}

Laurent, Antoine ^{[1
]}

Morin, Emmanuel ^{[5
]}

Esteve, Yannick ^{[3
]}

机构：

[1] Le Mans Univ, LIUM, Le Mans, France

[2] Univ Paris Saclay, CNRS, LIMSI, F-91400 Orsay, France

[3] Avignon Univ, LIA, Avignon, France

[4] McGill Univ, Montreal, PQ, Canada

[5] Univ Nantes, LS2N, Nantes, France

来源：

2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING | 2020年

关键词：

Spoken language understanding; end-to-end system; error analysis; neural network;

D O I：

10.1109/icassp40776.2020.9054455

中图分类号：

O42 [声学];

学科分类号：

070206 ; 082403 ;

摘要：

This paper presents a qualitative study of errors produced by an end-to-end spoken language understanding (SLU) system (speech signal to concepts) that reaches state of the art performance. Different studies are proposed to better understand the weaknesses of such systems: comparison to a classical pipeline SLU system, a study on the cause of concept deletions (the most frequent error), observation of a problem in the capability of the end-to-end SLU system to segment correctly concepts, analysis of the system behavior to process unseen concept/value pairs, analysis of the benefit of the curriculum-based transfer learning approach. Last, we proposed a way to compute embeddings of sub-sequences that seem to contain relevant information for future work.

引用

页码：8514 / 8518

页数：5

共 50 条

[1] TOWARDS END-TO-END SPOKEN LANGUAGE UNDERSTANDING
Serdyuk, Dmitriy
Wang, Yongqiang
Fuegen, Christian
Kumar, Anuj
Liu, Baiyang
Bengio, Yoshua
2018 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2018, : 5754 - 5758
[2] A Streaming End-to-End Framework For Spoken Language Understanding
Potdar, Nihal
Avila, Anderson R.
Xing, Chao
Wang, Dong
Cao, Yiran
Chen, Xiao
PROCEEDINGS OF THE THIRTIETH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2021, 2021, : 3906 - 3914
[3] Semantic Complexity in End-to-End Spoken Language Understanding
McKenna, Joseph P.
Choudhary, Samridhi
Saxon, Michael
Strimel, Grant P.
Mouchtaris, Athanasios
INTERSPEECH 2020, 2020, : 4273 - 4277
[4] WhiSLU: End-to-End Spoken Language Understanding with Whisper
Wang, Minghan
Li, Yinglu
Guo, Jiaxin
Qiao, Xiaosong
Li, Zongyao
Shang, Hengchao
Wei, Daimeng
Tao, Shimin
Zhang, Min
Yang, Hao
INTERSPEECH 2023, 2023, : 770 - 774
[5] End-to-End Spoken Language Understanding Without Full Transcripts
Kuo, Hong-Kwang J.
Tuske, Zoltan
Thomas, Samuel
Huang, Yinghui
Audhkhasi, Kartik
Kingsbury, Brian
Kurata, Gakuto
Kons, Zvi
Hoory, Ron
Lastras, Luis
INTERSPEECH 2020, 2020, : 906 - 910
[6] End-to-End Spoken Language Understanding for Generalized Voice Assistants
Saxon, Michael
Choudhary, Samridhi
McKenna, Joseph P.
Mouchtaris, Athanasios
INTERSPEECH 2021, 2021, : 4738 - 4742
[7] Exploring Transfer Learning For End-to-End Spoken Language Understanding
Rongali, Subendhu
Liu, Beiye
Cai, Liwei
Arkoudas, Konstantine
Su, Chengwei
Hamza, Wael
THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 13754 - 13761
[8] End-to-End Neural Transformer Based Spoken Language Understanding
Radfar, Martin
Mouchtaris, Athanasios
Kunzmann, Siegfried
INTERSPEECH 2020, 2020, : 866 - 870
[9] Privacy-Preserving End-to-End Spoken Language Understanding
Wang, Yinggui
Huang, Wei
Yang, Le
PROCEEDINGS OF THE THIRTY-SECOND INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, IJCAI 2023, 2023, : 5224 - 5232
[10] IN PURSUIT OF BABEL - MULTILINGUAL END-TO-END SPOKEN LANGUAGE UNDERSTANDING
Mueller, Markus
Choudhary, Samridhi
Chung, Clement
Mouchtaris, Athanasios
Kunzmann, Siegfried
2021 IEEE AUTOMATIC SPEECH RECOGNITION AND UNDERSTANDING WORKSHOP (ASRU), 2021, : 1042 - 1049

← 1 2 3 4 5 →