EXPERIMENT1AUDIOPROCESSING.docVIP

EXPERIMENT 1 AUDIO PROCESSING (3 hours) Aims To learn the procedures of digitizing sound, such as: Sampling Quantization Encoding To learn the significance of sampling and quantization for the audio quality. To learn how to edit and analyze audio. Basic Theory 2.1 Sound Fundamentals Sound is a physical phenomenon produced by the vibration of matter. Vibration of matter induces the oscillation of the air pressure around the matter. Then, the pressure variation propagates in a wave-like motion, and when it arrives at our ear, we hear the sound. Without air there is no sound, for example, in space. Since sound is a pressure wave, it takes on continuous values, as opposed to digitized ones with a finite range. Nevertheless, if we wish to use a digital version of sound waves, we must form digitized representations of audio information. The two most important parameters used to characterize the sound waveform are frequency and amplitude. Frequency is the reciprocal of period and represents the number of period within one second. The unit of frequency is Hz. The frequency range of speech is from 300Hz to 3400Hz. On the other hand, amplitude represents the loudness of sound. 2.2 Digitization of Sound As we know, sound is continuous analog signal, and can not be processed by computer directly. In computer, sound is represented by a series of digital number. We call the procedure of converting the analog sound waveform into discrete digital number as digitization of sound. For the sake of efficiency, these digital numbers should be integers. To fully digitize, the analog sound signal must be sampled in both time dimension and amplitude dimension, which is referred to as sampling and quantization respectively, as depicted in Fig.1. Furthermore, filtering and encoding are also important to digitization of sound. Prior to sampling, the analog sound signal should be filtered to remove the unwanted frequency by using a band pass filter. After quantization, encoding is performed to