Efficiency of End-to-End Speech Recognition for Languages with Scarce Resources

被引：0
|
作者：
Rudzionis, Vytautas [1 ]
Malukas, Ugnius [2 ]
Lopata, Audrius [2 ]
机构：
[1] Vilnius Univ, Kaunas Fac, Muitines 8, Kaunas, Lithuania
[2] Kaunas Univ Technol, Studentu 50, Kaunas, Lithuania
来源：
INFORMATION AND SOFTWARE TECHNOLOGIES, ICIST 2022 | 2022年 / 1665卷
关键词：
Speech recognition; Hybrid methods; Machine learning; End-to-end speech recognition;
D O I：
10.1007/978-3-031-16302-9_20
中图分类号：
TP [自动化技术、计算机技术];
学科分类号：
0812 ;
摘要：
Modern deep learning based speech recognition methods allow for achieving phenomenal speech recognition accuracy. But it requires enormous amounts of data to train such systems to achieve high recognition accuracy. Many less widely spoken languages simply do not possess the necessary amounts of speech corpora. The paper presents attempts to evaluate DeepSpeech-based speech recognition efficiency with the limited amounts of training data available and the ways to improve the accuracy. The experiments showed that the accuracy of DeepSpeech2 recognizer with about 100 h of speech corpora used for training is quite modest but the application of simple grammatical constraints allowed to reduce the word error rate to 23-25%.
引用
收藏
页码：259 / 264
页数：6
相关论文
共 50 条
  • [21] End-to-End Speech Recognition For Arabic Dialects
    Nasr, Seham
    Duwairi, Rehab
    Quwaider, Muhannad
    ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, 2023, 48 (08) : 10617 - 10633
  • [22] TOWARDS END-TO-END UNSUPERVISED SPEECH RECOGNITION
    Liu, Alexander H.
    Hsu, Wei-Ning
    Auli, Michael
    Baevski, Alexei
    2022 IEEE SPOKEN LANGUAGE TECHNOLOGY WORKSHOP, SLT, 2022, : 221 - 228
  • [23] An Overview of End-to-End Automatic Speech Recognition
    Wang, Dong
    Wang, Xiaodong
    Lv, Shaohe
    SYMMETRY-BASEL, 2019, 11 (08):
  • [24] TRIGGERED ATTENTION FOR END-TO-END SPEECH RECOGNITION
    Moritz, Niko
    Hori, Takaaki
    Le Roux, Jonathan
    2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2019, : 5666 - 5670
  • [25] Performance Monitoring for End-to-End Speech Recognition
    Li, Ruizhi
    Sell, Gregory
    Hermansky, Hynek
    INTERSPEECH 2019, 2019, : 2245 - 2249
  • [26] End-to-End Speech Recognition and Disfluency Removal
    Lou, Paria Jamshid
    Johnson, Mark
    FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, EMNLP 2020, 2020, : 2051 - 2061
  • [27] End-to-end Korean Digits Speech Recognition
    Roh, Jong-hyuk
    Cho, Kwantae
    Kim, Youngsam
    Cho, Sangrae
    2019 10TH INTERNATIONAL CONFERENCE ON INFORMATION AND COMMUNICATION TECHNOLOGY CONVERGENCE (ICTC): ICT CONVERGENCE LEADING THE AUTONOMOUS FUTURE, 2019, : 1137 - 1139
  • [28] END-TO-END MULTILINGUAL AUTOMATIC SPEECH RECOGNITION FOR LESS-RESOURCED LANGUAGES: THE CASE OF FOUR ETHIOPIAN LANGUAGES
    Abate, Solomon Teferra
    Tachbelie, Martha Yifiru
    Schultz, Tanja
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 7013 - 7017
  • [29] SPEECH ENHANCEMENT USING END-TO-END SPEECH RECOGNITION OBJECTIVES
    Subramanian, Aswin Shanmugam
    Wang, Xiaofei
    Baskar, Murali Karthick
    Watanabe, Shinji
    Taniguchi, Toru
    Tran, Dung
    Fujita, Yuya
    2019 IEEE WORKSHOP ON APPLICATIONS OF SIGNAL PROCESSING TO AUDIO AND ACOUSTICS (WASPAA), 2019, : 234 - 238
  • [30] Phonetically Induced Subwords for End-to-End Speech Recognition
    Papadourakis, Vasileios
    Mueller, Markus
    Liu, Jing
    Mouchtaris, Athanasios
    Omologo, Maurizio
    INTERSPEECH 2021, 2021, : 1992 - 1996
12345