Speaker Classification II pp 276-292 | Cite as
Selecting Representative Speakers for a Speech Database on the Basis of Heterogeneous Similarity Criteria
Abstract
In the context of the Neologos French speech database creation project, a general methodology was defined for the selection of representative speaker recordings. The selection aims at providing a good coverage in terms of speaker variability while limiting the number of recorded speakers. This is intended to make the resulting database both more adapted to the development of recently proposed multi-model methods and less expensive to collect.
The presented methodology proposes a selection process based on the optimization of a quality criterion defined in a variety of speaker similarity modeling frameworks. The selection can be achieved with respect to a unique similarity criterion, using classical clustering methods such as Hierarchical or K-Medians clustering, or it can combine several speaker similarity criteria, thanks to a newly developed clustering method called Focal Speakers Selection.
In this framework, four different speaker similarity criteria are tested, and three different speaker clustering algorithms are compared. Results pertaining to the collection of the Neologos database are also discussed.
Keywords
speech database minimization speaker selection speaker clustering optimal coverage multi-models speech and speaker recognition speech synthesisPreview
Unable to display preview. Download preview PDF.
References
- 1.Nagorski, A., Boves, L.: Steeneken: Optimal selection of speech data for automatic speech recognition systems. In: ICSLP, pp 2473–2476 (2002)Google Scholar
- 2.Lippmann, R.: Speech recognition by machines and humans. Speech Communication 22(1), 1–15 (1997)CrossRefGoogle Scholar
- 3.Iskra, D., Toto, T.: Speecon - speech databases for consumer devices: Database specification and validation. In: LREC, pp. 329–333 (2002)Google Scholar
- 4.Nakamura, A., Matsunaga, S., Shimizu, T., Tonomura, M., Sagisaka, Y.: Japanese speech databases for robust speech recognition. In: Proc. ICSLP 1996. Philadelphia, PA, vol. 4, pp. 2199–2202 (1996)Google Scholar
- 5.François, H., Boëffard, O.: Design of an optimal continuous speech database for text-to-speech synthesis considered as a set covering problem. In: Proc. Eurospeech 2001 (2001)Google Scholar
- 6.Krstulović, S., Bimbot, F., Boëffard, O., Charlet, D., Fohr, D., Mella, O.: Optimizing the coverage of a speech database through a selection of representative speaker recordings. Speech Communication 48(10), 1319–1348 (2006)CrossRefGoogle Scholar
- 7.Padmanabhan, M., Bahl, L., Nahamoo, D., Picheny, M.: Speaker clustering and transformation for speaker adaptation in speech recognition system. IEEE Transactions on Speech and Audio Processing 6(1), 71–77 (1998)CrossRefGoogle Scholar
- 8.Johnson, S., Woodland, P.: Speaker clustering using direct maximisation of the MLLR-adapted likelihood. In: ICSLP. vol. 5(98), pp. 1775–1779Google Scholar
- 9.Naito, M., Deng, L., Sagisaka, Y.: Speaker clustering for speech recognition using vocal tract parameters. Speech Communication 36(3-4), 305–315 (2002)MATHCrossRefGoogle Scholar
- 10.Gauvain, J., Lee, C.: Maximum a posteriori estimation for multivariate gaussian mixture observations of markov chains. IEEE Transactions on Speech and Audio Processing 2(2), 291–299 (1994)CrossRefGoogle Scholar
- 11.Reynolds, D.A., Quatieri, T., Dunn, R.: Speaker verification using adapted gaussian mixture models. Digital Signal Processing 10(1-3), 19–41 (2000)CrossRefGoogle Scholar
- 12.Ben, M., Blouet, R., Bimbot, F.: A Monte-Carlo method for score normalization in Automatic Speaker Verification using Kullback-Leibler distances. In: Proc. ICASSP 2002 (2002)Google Scholar
- 13.Duda, R.O., Hart, P.E., Stork, D.G.: Pattern Classification. John Wiley and Sons, New York (2001)MATHGoogle Scholar