Downloads: 139 | Views: 213

Research Paper | Electronics & Communication Engineering | India | Volume 8 Issue 4, April 2019

Speech Compression Using LPC-10 and VELPC Based on DCT Technique Compression

Aditi Kar ^[2] | Rasmi Ranjan Samantray ^[2]

Abstract: In this paper, we have presented the speech compression techniques i. e. Linear Predictive Coding and its modified version Voice Excited Linear Predictive Coding. Output of LPC-10 is compared with output of VELPC. Here instead of using plain VELPC we are using modified VELPC or we can say that VELPC based on Discrete Cosine Transform technique. The result shown here was obvious since VELPC is modified version of LPC and so its output quality is better as compared to output quality of LPC. LPC is one of the most powerful speech coding techniques and it is widely used for encoding speech signal at a very low bit rate. Pitch period estimation is the weakest link in the LPC method so this drawback is eliminated in VELPC technique. Thus, the bit rate however is somewhat increased in VELPC technique. Therefore Discrete Cosine Transform is used to compress the innovation signal in case of VELPC in order to reduce bit rate to some extent.

Keywords: Speech signal, Discrete Cosine Transform DCT, Linear Prediction Coding LPC, Voice Excited Linear Prediction Coding VELPC, Bit rate

Edition: Volume 8 Issue 4, April 2019,

Pages: 1196 - 1198