CELP DENGAN MENGGUNAKAN TRANSFORMASI EKSITASI PULSA BINER (TBPE)
<b>Abstract :</b><p align=\"justify\">Code Excited Linear Predictive (CELP) has been proven to be the promising coder for producing high quality speech at bit rate 4.8 kbps by vector quantizing the excitation sequence using a large stochastic codebook. Even though, a CELP...
Saved in:
| Main Author: | |
|---|---|
| Format: | Theses |
| Language: | Indonesia |
| Subjects: | |
| Online Access: | https://digilib.itb.ac.id/gdl/view/5321 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Institut Teknologi Bandung |
| Language: | Indonesia |
| Summary: | <b>Abstract :</b><p align=\"justify\">Code Excited Linear Predictive (CELP) has been proven to be the promising coder for producing high quality speech at bit rate 4.8 kbps by vector quantizing the excitation sequence using a large stochastic codebook. Even though, a CELP coder must run a huge computational to find the optimal excitation vector (search exhaustively) because it uses a large stochastic codebook (e.g. 1024 as in 4.8 kbps coding). <br />
<p align=\"justify\"> The reduction of the computational load for the CELP coder has been done by using efficient codebook structure, but it still searches the optimal excitation vector exhaustively. In this thesis I use Transformed Binary Pulse Excitation (TBPE) at the CELP coder and do some simulations because it significantly reduces the computational load. This computational load reduction can be achieved because the optimal excitation vector can be generated by using a smaller codebook with an appropriate transformation matrix. |
|---|