# Digital Signal Processing

631 Words3 Pages
ECE515EL: DSP LABORATORY Characteristics of Digital Audio Signals Digitally recorded audio signals can be characterized in terms of the sampling frequency (Fs) at which they are recorded, the number of bits used to represent the samples, and the number of audio channels recorded. Typical sampling frequencies used are 44100Hz (CD quality), 22050Hz (FM broadcast quality), 11050Hz (AM broadcast quality), and 8000Hz (Voice or Telephone quality). The samples are stored as 8-bit or 16-bit numbers. In terms of the number of audio channels, recorded audio signals may have 2 channels (stereo) or 1 channel only (mono). Activities: 1. Playback the following audio files which are recordings of a 1KHz tone. tone1K_16bit_44100Hz_mono.wav tone1K_16bit_8000Hz_mono.wav IMPORTANT: The filenames used the audio files for the exercises indicate the number of bits, sampling frequency, and the number of channels. For example, tone1K_16bit_44100Hz_mono.wav is a mono signal recorded at 44100Hz and uses 16-bit sample representation. (a) Is there a significant difference in the perceived quality of the tones played back from the files? (b) Check the file sizes. Which audio file is larger? Approximately, how much larger is it ? Explain the difference in the file sizes? (c) What is the theoretical minimum sampling frequency, according to the Nyquist sampling theorem, that could be used in recording the audio signal? If this theoretical minimum sampling frequency had been used, how much smaller would this file be compared to the 44100Hz file? (d) Which do you think is more practical: sampling the audio signal at 44100Hz or at 8000Hz? Explain your answer. 2. Playback the following audio files which contain recordings of a voice signal. voice_16bit_44100Hz_mono.wav voice_16bit_8000Hz_mono.wav (a) Is there a significant