Multidimensional Evaluation Platform for Call Center Speech Service Quality Based on Keyword Spotting

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 225)


Call center system has become an integral part of business operations and monitoring speech service quality is a vital work of the management in call center. Currently, main scoring approach is manual control mode. Its efficiency is not high and it is difficult to uniform standard. This paper presents an automatic multimode monitoring system for speech service quality of call center, and keyword spotting technology is applied to monitor the speech content automatically. In addition this paper designed a common set of evaluation algorithm and finally achieved the Multidimensional evaluation indicators to accurately and multidimensionally measure the operator’s speech QOS.


Call center Speech service quality Keyword spotting 


  1. 1.
    Wa QI (2004) Call center and its development. China Data Commun 1:45–68Google Scholar
  2. 2.
    HAN J, Liu XX, Pan JL (2120) A keyword spotting system for network and information security. China network and information security technology conference 22(03):56–61Google Scholar
  3. 3.
    Rose R and Paul D (1990) A hidden markrov model based keyword recognition system. 1990 IEEE ICASSP 1:129–132Google Scholar
  4. 4.
    Bazziand, Glass JR (2000) Modeling out-of vocabulary words for robust speech recognition. ICSLP 2000(1):1035–1038Google Scholar
  5. 5.
    Viterbi A (1967) Error bounds for convolutional codesandan asymptotieally optimum decoding algorithm. IEEE Trans Inf Theor 1:260–269CrossRefGoogle Scholar
  6. 6.
    Huang, Eng F, Wang, Hsiao Chuan (1994) A fast algorithm for large vocabulary keyword spotting application. J IEEE Trans Speech and Audio Process 1:449–452CrossRefGoogle Scholar
  7. 7.
    Evermann G (1999) Minimum word error rate decoding, vol 1, pp 45–67. Cambridge University, UKGoogle Scholar
  8. 8.
    Lu SQ, Sun CL (2006) Study of keyword recognition system for speech based on streaming media. J Beijing Inst Mach 1:89–94Google Scholar
  9. 9.
    Knill KM, Young SJ (1996) Fast implementation methods for Viterbi-based word-spotting. In: Proceedings of IEEE International Conference on Acoustics Speech and Signal Processing Atlanta USA 1:522-525Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.College of Computer Science & Information EngineeringZhejiang Gongshang UniversityHangzhouChina

Personalised recommendations