Speech Compression by using Adaptive Differential Pulse Code Modulation (ADPCM) technique With Microcontroller

Abstract

Compression is a process of reducing an input data (Speech Signal) bit stream into a small
bit size with high quality. Analog signal is a continuous signal takes more space to store the
data in memory devices with original size (Bit). All sensor data (Analog Data) stored in
computer with original size (Bit), but because of compression technique we store the same
data in reduced format we same quality. In compression the unwanted data is eliminate. The
main purpose of speech compression is to reduce the data bits for transmission of original
data from one place to other & store this data that maintaining the quality as same as
original signal. In this compression technique the analog to digital conversion (ADC)
process played important role, because of analog to digital conversion analog to digital
conversion (ADC) we get quantized sample signal. In that sample signal high correlation
property is present between the sampled speech signal. The Adaptive Delta Pulse Code
Modulation (ADPCM) techniques use the high correlation property of sampled data for
compression of signal. This algorithm cannot compress the sampled data as it. It takes the
difference between the predicted sample signal and actual sample signal then encode this
difference signal which is explained in details below. The Adaptive Delta Pulse Code
Modulation (ADPCM) methods have very efficient methods for the compression of signal by
reduction of number of bits per sample from original signal with maintaining the quality of
signal.
There are so many data compression technique available, but some technique algorithm
operation not gives actual quality of signal after compression. That type of technique is
called as lossy type algorithm. Because this type lossy algorithm the human ear cannot detect
the word. Human voice frequency ranges from 300 Hz to 3400 Hz. Adaptive Delta Pulse
Code Modulation (ADPCM) is a well known encoding scheme used for speech processing.
This project focuses on the simplification in the technique so that the hardware complexity
can be reduced for the portable speech compression & decompression devices.
In this project we used ARM controller which is heart of this project that contains 10-bit
channel analog to digital conversion (ADC) pin. Means we get the sample upto 1024. this
sample we have going uses for the Adaptive Delta Pulse Code Modulation (ADPCM)
algorithm. Also in ARM controller Digital to Analog coversion (DAC) pin has available to
check the compressed signal with original signal. Because of ARM controller we reduce the
circuitry. Also, we use the Digital signal oscilloscope (DSO) & personal computer to check
the behavioural of signal.
Because of compression we save the memory & transmission time with same quality. Also
when we want this data we check from stored from memory. In so many government offices,
private colleges, laboratory & industry requires to store the original data in computer or in
memory devices as it is, so many memory devices are required hence wastage of money is
take place. Because of compression by using adaptive differential pulse code modulation
technique with microcontroller we achieve the compression of signal with same qualit