Advances in Profile Assisted Voicemail Management

  • Konstantinos Koumpis
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3192)


Spoken audio is an important source of information available to knowledge extraction and management systems. Organization of spoken messages by priority and content can facilitate knowledge capture and decision making based on profiles of recipients as these can be determined by physical and social conditions. This paper revisits the above task and addresses a related data sparseness problem. We propose a methodology according to which the coverage of language models used to categorize message types is augmented with previously unobserved lexical information derived from other corpora. Such lexical information is the result of combining word classes constructed by an agglomerative clustering algorithm which follows a criterion of minimum loss in average mutual information. We subsequently generate more robust category estimators by interpolating class-based and voicemail word-based models.


automatic categorization speech recognition stochastic language models class-based clustering voicemail 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Moreno, P., Thong, J.M.V., Logan, B., Jones, G.J.F.: From multimedia retrieval to knowledge management. IEEE Computer 35, 58–66 (2002)Google Scholar
  2. 2.
    Hirschberg, J., Bacchiani, M., Hindle, D., Isenhour, P., Rosenberg, A., Stark, L., Stead, L., Whittaker, S., Zamchick, G.: SCANMail: Browsing and searching speech data by content. In: Proc. Eurospeech, Aalborg, Denmark (2001)Google Scholar
  3. 3.
    Koumpis, K., Ladas, C., Renals, S.: An advanced integrated architecture for wireless voicemail retrieval. In: Proc. 15th IEEE Intl. Conf. on Information Networking, Beppu, Japan, pp. 403–410 (2001)Google Scholar
  4. 4.
    Huang, J., Zweig, G., Padmanabhan, M.: Information extraction from voicemail. In: 39th Annual Meeting of Assoc. for Computational Linguistics, Toulouse, France (2001)Google Scholar
  5. 5.
    Ringel, M., Hirschberg, J.: Automated message prioritization: Making voicemail retrieval more efficient. In: Proc. Conf. on Human Factors in Computing Systems (Ext. Abstracts), Minneapolis, MN, USA, pp. 592–593 (2002)Google Scholar
  6. 6.
    Yang, Y.: An evaluation of statistical approaches to text categorization. Journal of Information Retrieval 1, 67–88 (1999)Google Scholar
  7. 7.
    Lewis, D.D., Schapire, R.E., Callan, J.P., Papka, R.: Algorithms for linear text classifiers. In: Proc. 19th annual Intl. ACM SIGIR Conf. on Research and Development in Information Retrieval, pp. 298–306 (1996)Google Scholar
  8. 8.
    Sebastiani, F.: Machine learning in automated text categorization. ACM Computing Surveys 34, 1–47 (2002)CrossRefMathSciNetGoogle Scholar
  9. 9.
    Teahan, W.J., Harper, D.J.: Using compression based language models for text categorization. In: Proc. Workshop on Language Modeling and Information Retrieval, Carnegie Mellon University, USA, pp. 83–88 (2001)Google Scholar
  10. 10.
    Peng, F., Schuurmans, D., Kaselj, V., Wang, S.: Automated authorship attribution with character level language models. In: Proc. 10th Conf. of European Chapter of Assoc. for Computational Linguistics, Budapest, Hungary, pp. 19–24 (2003)Google Scholar
  11. 11.
    Padmanabhan, M., Eide, E., Ramabhardan, G., Ramaswany, G., Bahl, L.: Speech recognition performance on a voicemail transcription task. In: Proc. IEEE ICASSP, Seattle, WA, USA, pp. 913–916 (1998)Google Scholar
  12. 12.
    Koumpis, K., Renals, S.: The role of prosody in a voicemail summarization system. In: Proc. ISCA Workshop on Prosody in Speech Recognition and Understanding, Red Bank, NJ, USA, pp. 87–92 (2001)Google Scholar
  13. 13.
    Cordoba, R., Woodland, P.C., Gales, M.J.F.: Improving cross task performance using MMI training. In: Proc. IEEE ICASSP, Orlando, FL, USA, vol. 1, pp. 85–88 (2002)Google Scholar
  14. 14.
    Koumpis, K.: Automatic categorization of voicemail transcripts using stochastic language models. In: Proc. 7th Int. Conf. on Text, Speech and Dialogue, Brno, Czech Republic (2004)Google Scholar
  15. 15.
    Charlet, D.: Speaker indexing for retrieval of voicemail messages. In: Proc. IEEE ICASSP, Orlando, FL, USA, vol. 1, pp. 121–124 (2002)Google Scholar
  16. 16.
    Gotoh, Y., Renals, S.: Statistical language modelling. In: Renals, S., Grefenstette, G. (eds.) Text and Speech Triggered Information Access, pp. 78–105. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  17. 17.
    Jelinek, F., Mercer, R.L., Bahl, L.R., Baker, J.K.: Perplexity - a measure of difficulty of speech recognition tasks. In: Proc. 94th Meeting Acoustical Society of America, Miami Beach, Florida, USA (1977)Google Scholar
  18. 18.
    Brown, P.F., Pietra, V.J.D., deSouza, P.V., Lai, J.C., Mercer, R.L.: Class-based n-gram models of natural language. Computational Linguistics 18, 467–479 (1992)Google Scholar
  19. 19.
    Chen, S., Goodman, J.: An empirical study of smoothing techniques for language modeling. Computer Speech and Language 13, 359–394 (1999)CrossRefGoogle Scholar
  20. 20.
    Carletta, J.: Assessing agreement on classification tasks: The kappa statistic. Computational Linguistics 22, 249–254 (1996)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Konstantinos Koumpis
    • 1
  1. 1.Vienna Telecommunications Research Center – ftwTech Gate ViennaViennaAustria

Personalised recommendations