Sciweavers

INTERSPEECH
2010

Efficient HMM-based estimation of missing features, with applications to packet loss concealment

12 years 10 months ago
Efficient HMM-based estimation of missing features, with applications to packet loss concealment
In this paper, we present efficient HMM-based techniques for estimating missing features. By assuming speech features to be observations of hidden Markov processes, we derive a minimum mean-square error (MMSE) solution. We increase the computational efficiency of HMM-based methods by downsampling underlying Markov models, and by enforcing symmetry in transitional probability matrices. When applied to features generally utilized in parametric speech coding, namely line spectral frequencies (LSFs), the proposed methods provide significant improvement over the baseline repetition scheme, in terms of Itakura-Saito distortion and peak SNR.
Bengt J. Borgström, Per Henrik Borgström
Added 18 May 2011
Updated 18 May 2011
Type Journal
Year 2010
Where INTERSPEECH
Authors Bengt J. Borgström, Per Henrik Borgström, Abeer Alwan
Comments (0)