Improved algorithms for phase prediction and frame interpolation in low bit rate sinusoidal coders

A number of improved algorithms for phase prediction and frame interpolation in the context of sinusoidal speech coding are presented. A minimum variance sinusoidal phase estimation scheme is proposed, where it is shown that reasonably accurate estimates for short-time sinusoidal phases corresponding to voiced frames are obtained. On the other hand, improved algorithms for interpolation of sine wave parameters are presented which result in further reduction in bit rate while preserving the subjective quality of the reproduced speech at low bit rates. The performance of the above algorithms was evaluated on a large speech database and the results of statistical analysis are provided. The proposed algorithms were successfully integrated into a 2.4 kbps sinusoidal coder, where speech of good quality, intelligibility, and naturalness has been obtained.