论文信息 - Speaker independent digit recognition with reference frame-specific distance measures

Speaker independent digit recognition with reference frame-specific distance measures

This work defines novel distance measures for speech recognition which: 1. Model the statistical interaction between adjacent speech frames, 2. Model the statistical characteristics of different speech sounds individually, 3. Are specifically designed to differentiate between confusable speech sounds. Speaker independent recognition tests performed on the Texas Instruments multi-dialect isolated digit data base give substitution rates as low as 0.6 % with a vocabulary of 11 digits.

G. Doddington | E. Bocchieri

[1] S. Levinson,et al. Considerations in dynamic time warping algorithms for discrete word recognition , 1978 .

[2] Aaron E. Rosenberg,et al. Considerations in dynamic time warping algorithms for discrete word recognition , 1978 .

[3] Louis C. W. Pols,et al. Real-Time Recognition of Spoken Words , 1971, IEEE Transactions on Computers.

[4] Gary E. Kopec,et al. Network-based isolated digit recognition using vector quantization , 1985, IEEE Trans. Acoust. Speech Signal Process..

[5] Julius T. Tou,et al. Pattern Recognition Principles , 1974 .

[6] R. G. Leonard,et al. A database for speaker-independent digit recognition , 1984, ICASSP.

[7] Aaron E. Rosenberg,et al. Speaker-independent recognition of isolated words using clustering techniques , 1979 .

[8] G. Doddington,et al. Speech recognition in the F-16 cockpit using principal spectral components , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.