PERIOD INFORMATION QUANTIZATION CODEBOOK OF THE SEGMENTAL SINUSOIDAL MODEL
| Main Author: | Setiawan, Florentinus Budi |
|---|---|
| Format: | Article PeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
Unika Soegijapranata
|
| Subjects: | |
| Online Access: |
http://repository.unika.ac.id/29695/1/SKI%202014%20Period%20Informaton%20Quantization.pdf http://repository.unika.ac.id/29695/ |
Daftar Isi:
- Segmental sinusoidal model is an approximation method based on sinusoidal model for speech signal, especially for periodic part. The periodic signal can be decomposed by infinite sinusoidal signal with combination of amplitude, frequency and phase. Quantization is a method to code or compress a speech signal. The proposed quantization method in this paper is sampling signal at minimum and maximum part over one certain block. Parameters of speech signal are detected at its peaks, both positive and negative peaks. The useful parameters are peaks and period between consecutive peaks. To obtain the lower size of signal parameter, a look-up table is needed Signal parameters are quantized into vector quantization form, so that codebook has to be generated in order to indexing of the vector quantization. In this paper, we show the experimental results, consist of codebook design and its performance based on the reconstruction of signal. Reconstruction process is realized based on periods and peaks information. The experimental results show that this is a correlation between the size of codebook and performance of reconstructed signal