Comparative Analysis of Korean Continuous Speech Recognition Accuracy by Application Field of Cloud-Based Speech Recognition Open API 


Vol. 45,  No. 10, pp. 1793-1803, Oct.  2020
10.7840/kics.2020.45.10.1793


PDF
  Abstract

Speech recognition has significantly improved performance through the application of deep learning technology and the emergence of cloud computing. The improved speech recognition has been applied in various fields such as vehicles, robot, healthcare, and call center. This paper compares the performance of continuous Korean speech recognition by application field for the cloud-based speech recognition Open API. The experiment was conducted with 7 domestic and foreign cloud companies. Korean continuous speech data was used by news data from three domestic broadcasters. The collection methods were divided into a total of 10 fields and collected 15 sentences by sector, a total of 150 sentences. As a result of the experiment, the overall speech recognition accuracy by field was the highest in Kakao and the lowest in IBM. By field, Kakao showed good performance in 6 fields, Amazon and Microsoft in 2 fields, and ETRI in 1 field. As a result of the experiment, it was confirmed that the speech recognition engine supported by the cloud computing company exhibits excellent performance in a specific field. This study is hoped to contribute to improving the performance of speech recognition engines for companies that support cloud-based speech recognition open APIs. In addition, for speech recognition developers, it is expected that it will help to select the most suitable voice recognition open API for the application field in developing an applied voice recognition system.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

H. Yoo, M. Kim, S. Park, K. Kim, "Comparative Analysis of Korean Continuous Speech Recognition Accuracy by Application Field of Cloud-Based Speech Recognition Open API," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 10, pp. 1793-1803, 2020. DOI: 10.7840/kics.2020.45.10.1793.

[ACM Style]

Hyun-Jae Yoo, Myung-Wha Kim, Sang-Kil Park, and Kwang-Yong Kim. 2020. Comparative Analysis of Korean Continuous Speech Recognition Accuracy by Application Field of Cloud-Based Speech Recognition Open API. The Journal of Korean Institute of Communications and Information Sciences, 45, 10, (2020), 1793-1803. DOI: 10.7840/kics.2020.45.10.1793.

[KICS Style]

Hyun-Jae Yoo, Myung-Wha Kim, Sang-Kil Park, Kwang-Yong Kim, "Comparative Analysis of Korean Continuous Speech Recognition Accuracy by Application Field of Cloud-Based Speech Recognition Open API," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 10, pp. 1793-1803, 10. 2020. (https://doi.org/10.7840/kics.2020.45.10.1793)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880