Data augmentation and language model adaptation using singular value decomposition

被引：0
|
作者：
Béchet, F [1 ]
De Mori, R [1 ]
Janiszek, D [1 ]
机构：
[1] Univ Avignon, CERI, LIA, F-84911 Avignon 9, France
来源：
PATTERN RECOGNITION LETTERS | 2004年 / 25卷 / 01期
关键词：
statistical language modeling; singular value decomposition; data augmentation; automatic speech recognition;
D O I：
10.1016/j.patrec.2003.08.008
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
A new augmentation method for counts to be used in language modeling is presented. It is based on word representations in a reduced space obtained with Singular Value Decomposition. A contribution to a count for a linguistic event x is obtained from the counts of observed events smoothed with a function of their distance from x. Experimental results on a spoken dialogue corpus show the performance of the proposed method, combined with maximum a posteriori probability adaptation, in terms of word error rate reduction. (C) 2003 Elsevier B.V. All rights reserved.
引用
收藏
页码：15 / 19
页数：5
相关论文
共 50 条
  • [1] Data augmentation and language model adaptation
    Janiszek, D
    De Mori, R
    Bechet, E
    [J]. 2001 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS I-VI, PROCEEDINGS: VOL I: SPEECH PROCESSING 1; VOL II: SPEECH PROCESSING 2 IND TECHNOL TRACK DESIGN & IMPLEMENTATION OF SIGNAL PROCESSING SYSTEMS NEURALNETWORKS FOR SIGNAL PROCESSING; VOL III: IMAGE & MULTIDIMENSIONAL SIGNAL PROCESSING MULTIMEDIA SIGNAL PROCESSING - VOL IV: SIGNAL PROCESSING FOR COMMUNICATIONS; VOL V: SIGNAL PROCESSING EDUCATION SENSOR ARRAY & MULTICHANNEL SIGNAL PROCESSING AUDIO & ELECTROACOUSTICS; VOL VI: SIGNAL PROCESSING THEORY & METHODS STUDENT FORUM, 2001, : 549 - 552
  • [2] Tensor Decomposition of Biometric Data using Singular Value Decomposition
    Mistry, Nirav
    Tanwar, Sudeep
    Tyagi, Sudhanshu
    Singh, Pradeep Kr
    [J]. 2018 FIFTH INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED AND GRID COMPUTING (IEEE PDGC), 2018, : 833 - 837
  • [3] Compression of magnetohydrodynamic simulation data using singular value decomposition
    del-Castillo-Negrete, D.
    Hirshman, S. P.
    Spong, D. A.
    D'Azevedo, E. F.
    [J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2007, 222 (01) : 265 - 286
  • [4] ECG data compression using truncated singular value decomposition
    Wei, JJ
    Chang, CJ
    Chou, NK
    Jan, GJ
    [J]. IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, 2001, 5 (04): : 290 - 299
  • [5] ECG Data Encryption Then Compression Using Singular Value Decomposition
    Liu, Ting Yu
    Lin, Kuan Jen
    Wu, Hsi Chun
    [J]. IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2018, 22 (03) : 707 - 713
  • [6] Analysis of Singular Value Decomposition using High Dimensionality Data
    On, Fatin Raihana
    Jailani, Rozita
    Hassan, Siti Lailatul
    Tahir, Nooritawati Md
    [J]. 2015 IEEE 11TH INTERNATIONAL COLLOQUIUM ON SIGNAL PROCESSING & ITS APPLICATIONS (CSPA 2015), 2015, : 186 - 191
  • [7] Faster Imputation Using Singular Value Decomposition for Sparse Data
    Phuc Nguyen
    Tran, Linh G. H.
    Le, Bao H.
    Nguyen, Thuong H. T.
    Thu Nguyen
    Nguyen, Hien D.
    Nguyen, Binh T.
    [J]. INTELLIGENT INFORMATION AND DATABASE SYSTEMS, ACIIDS 2023, PT I, 2023, 13995 : 135 - 146
  • [8] Singular value decomposition based recommendation using imputed data
    Yuan, Xiaofeng
    Han, Lixin
    Qian, Subin
    Xu, Guoxia
    Yan, Hong
    [J]. KNOWLEDGE-BASED SYSTEMS, 2019, 163 : 485 - 494
  • [9] Using singular value decomposition in conjunction with data assimilation procedures
    Dimitriu, Gabriel
    [J]. Numerical Methods and Applications, 2007, 4310 : 435 - 442
  • [10] DATA RECONCILIATION AND THE SINGULAR VALUE DECOMPOSITION
    Mitsas, Christos L.
    [J]. XIX IMEKO WORLD CONGRESS: FUNDAMENTAL AND APPLIED METROLOGY, PROCEEDINGS, 2009, : 2374 - 2377
12345