Optimality and Complexity Considerations

  • V. Ramasubramanian
  • Harish Doddala
Chapter
Part of the SpringerBriefs in Electrical and Computer Engineering book series (BRIEFSELECTRIC)

Abstract

We propose a low complexity unit-selection algorithm for ultra low bit-rate speech coding based on a first-stage n-best pre-quantization lattice and a second-stage run-length constrained Viterbi search to efficiently approximate the complete search space of the fully-optimal 1-pass DP based unit-selection algorithm described in the previous chapter. By this, the n-best low complexity algorithm dealt with here, approaches near-optimality with increasing n, in terms of rate-distortion performance while having highly reduced complexity. The sub-optimal segmental unit-selection algorithm of Lee and Cox described in Chap. 3 is a 1-best special case of the algorithm proposed here, with the proposed n-best lattice based algorithm generalizing to a larger search space and hence having a significantly improved rate-distortion performance towards the fully optimal 1-pass DP unit-selection performance.

References

  1. [HR08]
    D. Harish, V. Ramasubramanian, Comparison of segment quantizers: VQ, MQ, VLSQ and Unit-selection algorithms for ultra low bit-rate speech coding, in Proceedings of ICASSP ’08, Las Vegas, Mar 2008, pp. 4773–4776Google Scholar
  2. [LC02]
    K.S. Lee, R.V. Cox, A segmental speech coder based on a concatenative TTS. Speech Commun. 38, 89–100 (2002)CrossRefMATHGoogle Scholar
  3. [RH07]
    V. Ramasubramanian, D. Harish, An optimal unit-selection algorithm for ultra low bit-rate speech coding, in Proceedings of ICASSP ’07, Hawaii, Apr 2007, pp. IC-541–IC-544Google Scholar
  4. [RH06]
    V. Ramasubramanian, D. Harish, An unified unit-selection framework for ultra low bit-rate speech coding, in Proceedings of Interspeech ’06, Pittsburgh, Sept 2006, pp. 217–220Google Scholar
  5. [RH08]
    V. Ramasubramanian, D. Harish, Low complexity near-optimal unit-selection algorithm for ultra low bit-rate speech coding based on N-best lattice and Viterbi decoding, in Proceedings of Interspeech ’08, Brisbane, Sept 2008, p. 44Google Scholar
  6. [RSM82b]
    S. Roucos, R. Schwartz, J. Makhoul, Segment quantization for very-low-rate speech coding, in Proceedings of ICASSP, vol. 3, Paris, France, 1982, pp. 1565–1568Google Scholar
  7. [RSM83]
    S. Roucous, R.M. Schwartz, J. Makhoul, A segment vocoder at 150 b/s, Proceedings of ICASSP ’83, Boston, 1983, pp. 61–64Google Scholar

Copyright information

© The Author 2015

Authors and Affiliations

  • V. Ramasubramanian
    • 1
  • Harish Doddala
    • 2
  1. 1.PES Institute of Technology – Bangalore South CampusBangaloreIndia
  2. 2.OracleRedwood ShoresUSA

Personalised recommendations