Chapter 5 Fundamentals of Audio

Fundamentals of Digital Sound
 Introduction to sound.
 Multimedia system sound.
 Digital audio.
 MIDI audio.
 Audio file formats.

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)
1
Introduction to Sound
 Sound is a form of energy, caused by molecules vibrating
 Vibrations in the air create waves of pressure perceived as
sound.
 Sound waves vary in sound pressure level (amplitude) and in
frequency or pitch.
 Sound pressure levels measured in decibels (dB).
 Humans hear sound over a very broad range.

2
Introduction . . . (cont’d)
 Sound comprises the spoken word, voices, music and even
noise. It is a complex relationship involving:

 a vibrating object (sound source)
 a transmission medium (usually air)
 a receiver (ear) and;
 a preceptor (brain).

3
Introduction . . . (cont’d)
Sound is described in terms of two characteristics:
– Frequency (or pitch)
– Amplitude (or loudness)

4
Frequency of Sound
 It is measure of how many vibrations occur in one second.
 Measured in Hertz (abbreviation Hz) and directly

corresponds to the pitch of a sound.
 The more frequent vibration occurs the higher the pitch of
the sound.
Low Pitch High Pitch

5
Frequency of Sound
 People can hear from 20 Hz to 20,000 Hz (20 kHz)
 Sounds below 20 Hz are infrasonic.
 Sounds above 20 kHz are ultrasonic.

6
Amplitude
 It is the maximum displacement of a wave from an equilibrium
position.
 Loud sounds have a large amplitude. (The louder the sound, the
more energy it has).
Quiet Loud
Low amplitude High Amplitude

7
Types of Sound
Analog sound
 Sound itself is a continuous wave; it is an analog signal
.
 It is the true representation of the sound at the moment
it was recorded.
 what the microphone receives is exactly what's written
onto the vinyl disk or cassette.

8
Types of Sound
Digital sound
 It is not a recording of the actual sound, but rather a
combination of binary code.

 The binary code is arranged in a specific pattern
informing the computer how to recreate the sound
itself.

9
Analog to Digital Conversion
 Why Digitize?
 Microphones, video cameras produce analog signals (continuous

valued voltages)
 To store audio or video data into a computer, it must digitize it by
converting it into a stream of numbers.
Digitizing
 the process of converting an analog signal to a

digital one.

10
Analog to Digital . . . (Cont’d)

11
Digitizing
 the process of converting an analog signal to a digital one.
 Analog signal is sampled. Digitized audio is sampled audio.
 Sampling is the process of taking periodic measurements of
the continuous signal.

 Samples are taken at regular time interval, i.e. every T
seconds.
 Many times each second, the analog signal is sampled. How
often these samples are taken is referred to as sampling rate.

 The amount of information stored about each sample is
referred to as sample size.
12 IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 The higher the sampling rate, the more the measurements are
taken (better quality).
 The lower the sampling rate, the lesser the measurements are
taken (low quality)
High Sampling Rate
Low Sampling Rate

13

 Sampling: Divide the horizontal axis (time) into discrete pieces
 Quantization: Divide the vertical axis (signal strength - voltage)
into pieces.
 8-bit quantization divides the vertical axis into 256 levels (.
 16-bit gives you 65536 levels ().
 Lower the quantization, lower the quality of the sound.
 If the scale used for the vertical axis is linear we say its linear
quantization.
 If its logarithmic then we call it non-linear.
14

15
Analog vs. Digital Sound
Analog Digital
Signal Continuous signal which Discrete time signals
represents physical generated by digital
measurements. modulation.
Waves Denoted by sine waves Denoted by square waves
Information Uses continuous range of Uses discrete or
Representation values discontinuous values
Example Human voice in air, analog Computers, CDs, DVDs,
electronic devices. and other digital electronic
devices.

16
Analog vs. Digital . . . (Cont’d)
Analog Digital
Technology Records waveforms as Samples analog waveforms into a
they are. limited set of numbers and records
them.
Data Subjected to deterioration Can be noise-immune without

transmissions by noise during deterioration during transmission
transmission and and write/read cycle.
write/read cycle.
Response to More likely to get affected Less affected since noise response
Noise reducing accuracy are analog in nature
Flexibility Analog hardware is not Digital hardware is flexible in
flexible. implementation.
Analog vs. Digital . . . (Cont’d)
Analog Digital
Uses Can be used in analog Best suited for Computing and
devices only. Best suited for digital electronics.
audio and video
transmission.
Bandwidth Analog signal processing There is no guarantee that digital
can be done in real time and signal processing can be done in
consumes less bandwidth. real time and consumes more
bandwidth to carry out the same
information.
Memory Stored in the form of wave Stored in the form of binary bit
signal

18
 Common Sampling Rates
 8KHz: used for telephone
 11.025 KHz: Speech audio
 22.05 KHz: Low Grade Audio (WWW Audio, AM Radio)
 44.1 KHz: CD Quality audio
 Sample Resolution/Sample Size
 Each sample can be measured to a certain degree of accuracy.
 The accuracy is dependent on the number of bits used to
represent the amplitude (sample resolution).

19

How do we store each sample value (quantized value)?
 8 Bit Value (0-255)
 16 Bit Value (Integer) (0-65535)
 The amount of memory required to store t seconds long
• is the sampling Frequency

• is the duration of time in seconds
 For 8 bit resolution, mono recording
 For 8 bit resolution, stereo recording
 For 16 bit resolution, mono recording
 For 16 bit resolution, stereo recording

20
Graphic/Image . . . (Cont’d)
 Exercise
 Someone sampled audio for 10 seconds. How much storage space is required
if
 22.05 KHz sampling rate is used, and 8 bit resolution with mono
recording? (Ans. 220.5KB )

 44.1 KHz sampling rate is used, and 8 bit resolution with mono recording?
(Ans. 441KB )
 44.1 KHz sampling rate is used, 16 bit resolution with stereo recording?
(Ans. 1764KB )
 11.025 KHz sampling rate, 16 bit resolution with stereo recording?
(Ans. 441KB )

21
Musical Instrument Digital Interface (MIDI)

22
MIDI (Cont’d)
 MIDI is protocol that enables to communicate computer,
synthesizers, keyboards, and other musical devices with each

other.
 Setup:
 MIDI OUT of synthesizer is connected to MIDI IN of
sequencer.
 MIDI OUT of sequencer is connected to MIDI IN of
synthesizer and "through" to each of the additional

sound modules.

23
MIDI (Cont’d)
 Working:
 During recording, the keyboard equipped synthesizer is used
to send MIDI message to the sequencer, which records them.

 During play back, messages are sent out from the sequencer
to the sound modules and the synthesizer which will play

back the music.

24
Components of MIDI
 Synthesizer
 It is a sound generator (various pitch, loudness, tone)
 A good synthesizer often has microprocessor, keyboard,
control panel, memory, etc.

 Sequencer
 A stand-alone unit or a software program for a personal
computer.
 It can be a storage server for MIDI data.
 Nowadays, it is more music editor software.
 It has one or more MIDI in and MIDI out.

25
Components of MIDI
 Track
 Track in sequencer is used to organize the recording.

 Track can be turned on or off on recording or playing back.
 Channel
 Used to separate information in a MIDI system.

 There are 16 channel in MIDI system.
 Channel numbers are coded in each MIDI message.
 Timbre
 The quality of sound

 Multitimbral:- capable of playing many different sounds at the same
time (piano, flute, drum, . . ., etc)

Editing Operations in Digital Audio
 Trimming: Removing dead air or blank space and so on. Done
using commands like cut , clear , Erase, Silence.

 Splicing and Assembly: Removal of extraneous noise , touch up,
creating a longer recording by mixing many small recordings.

 Volume adjustments: when we combine one or more recording
with different volume level, they must be made to run with a

consistent volume level. We need to normalize it with a sound
editor to a particular level.

27
Editing Operations in . . . (Cont’d)
 Format Conversion: While saving different formats are
made available for sounds. Data may be lost during

conversion. MP3,MP4
 Resampling and down sampling: if we have sound recorded
and edited the sound at 16 bit sampling rate and using lower
rates then RS & DS is necessary, as we may save lots of disk
space.
 Fade in & Fade out: useful for sections that requires this
effect. This enveloping helps to smooth out a beginning and

end of the sound file. Format conversion
28
 Equalization: allow us to modify a recordings frequency
content so that it sounds brighter or darker

 Time Stretching: allows to alter the length of the file .
Useful but may degrade audio quality of the file.

 Digital Sound Processing: allows signals with
reverberation, multi tap delay, chorus, flange and other
special effects using DSP routines.

29
 Reversing Sound: Reversing or a part of a recording.

 Multiple Tracks: Being able to edit and combine multiple
tracks, merge tracks and then export them in a final mix to a
single audio file is important.

30
MIDI vs. Audio sound
Comparison MIDI Digitized Audio
Representation Shorthand representation Sampling is done to
of music stored in convert the data into
numeric form digital form
Device Device dependent Device Independent
Independence
File Size 200 to 1000 times Larger File Size
smaller than Digitized
Audio
Memory Less storage space Large Storage space
Requirement
Edit Option Easily editable and all the Difficult to edit.
information retainable

31
MIDI Vs Audio sound
Comparison MIDI Digitized Audio

Quality Better when played on high Not so better
quality MIDI device
Playback Does not have consistent Consistent
playback quality playback quality
Analogy Vector Graphics Bitmap Images
Ease to Must have knowledge Does not require
incorporate much knowledge

32
Audio File Format
 It is a file format for storing digital audio data on a computer
system.
 This data can be stored uncompressed, or compressed to reduce
the file size.

 It can be a raw bit-stream, but it is usually a container format
or an audio data format with defined storage layer.

 The three major groups of audio file formats.
 Uncompressed audio formats.
 Formats with lossless compression.
 Formats with lossy compression.

Audio File Format (Cont’d)
 Uncompressed audio formats,
 such as WAV, AIFF, AU or raw header-less PCM;
 Formats with lossless compression
 such as FLAC, Monkey's Audio (filename extension APE),
WavPack (filename extension WV), TTA, ATRAC Advanced

Lossless, Apple Lossless (filename extension m4a), MPEG-4 SLS,
MPEG-4 ALS, MPEG-4 DST, Windows Media Audio Lossless
(WMA Lossless), and Shorten (SHN).
 Formats with lossy compression
 such as MP3, Vorbis, Musepack, AAC, ATRAC and Windows Media
Audio Lossy (WMA lossy)).

34
 .mid MIDI (Musicl Instrument Digital Interface)

 .wav Waveform Extension
 .aif Audio Interchange Format
 .mp2 .mp3 Motion Picture Expert Group Audio - also
referred to as MPEG-1 layer-2 or MPEG-1
Layer-3
 .ra .ram .rpm Real Audio

35
 .MID
 Wide support in many browsers with no need for a plugin.
 Have very good sound quality, but this can vary somewhat
with the quality of the sound card.

 Very small file size for a lengthy selection.
 The files are instrumental only.
 The files can not be recorded.
 They must be synthesized on a computer with special
hardware and software.

36
 WAV
 Very good sound quality.
 Widely supported in many browsers with no need for a
plugin.
 You can record your own .wav files from a CD, tape,
microphone, etc.
 The very large file sizes severely limit the length of the
sound clips that you can use on your Web pages.

37
 .AIF
 Very good sound quality.

 Widely supported in many browsers with no need for a
plugin.
 You can record your own .aif files from a CD, tape,
microphone, etc.
 The very large file sizes severely limit the length of the
sound clips that you can use on your Web pages.

38
 .MP2/.MP3/.Mp4
 It is a compressed format which makes the sound files smaller.

 The sound quality is very good.
 If an MP3 file is recorded and compressed properly, the quality
can rival that of an actual CD.

 The file size is still larger than a Real Audio file and a whole
song would still take quite awhile to download over a normal
phone line connection.

39
 .RA/.RAM/.RPM
 Very high degree of compression with smaller file sizes

than .mp2 or .mp3.
 Whole songs files are reasonable to download.
 The files can be "streamed" from a normal Web server
without any special software so that the audience can begin
listening to the sound before the file has completely
downloaded.
 Whole songs will start playing within seconds over normal
phone line connections.
40
 .RA/.RAM/.RPM (Cont’d)
 The sound quality is poorer than .mp2/ .mp3 files, but the new
G2 player and encoder have increased the quality considerably.

 Your audience must download and install a helper application
or plugin in order to appreciate the excellent quality of the new
G2 standard.
 However, browser manufacturers are beginning to include a
plug-in for the older version 5 player with the browser
downloads.

41
Popular Media Players

42
 MatLab Exercise on reading Audio files
clc
[y, Fs] = audioread('HBD2.mp3');
player = audioplayer(y, Fs)
play(player);
pause(player);
resume(player);
stop(player);

43

Chapter 5 Fundamentals of Audio

Uploaded by

Copyright:

Available Formats

Chapter 5 Fundamentals of Audio

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 5 Fundamentals of Audio

Uploaded by

Copyright:

Available Formats

Fundamentals of Digital Sound

 Multimedia system sound.

 Audio file formats.

 Sound is a form of energy, caused by molecules vibrating

 Vibrations in the air create waves of pressure perceived as

 Humans hear sound over a very broad range.

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

noise. It is a complex relationship involving:

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

– Frequency (or pitch)

– Amplitude (or loudness)

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 Measured in Hertz (abbreviation Hz) and directly

Low Pitch High Pitch

 Sounds below 20 Hz are infrasonic.

 Sounds above 20 kHz are ultrasonic.

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

more energy it has).

Low amplitude High Amplitude

onto the vinyl disk or cassette.

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

combination of binary code.

informing the computer how to recreate the sound

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 Microphones, video cameras produce analog signals (continuous

 the process of converting an analog signal to a

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 Analog signal is sampled. Digitized audio is sampled audio.

 Sampling is the process of taking periodic measurements of

the continuous signal.

often these samples are taken is referred to as sampling rate.

referred to as sample size.

12 IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

Low Sampling Rate

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 Quantization: Divide the vertical axis (signal strength - voltage)

 16-bit gives you 65536 levels ().

 Lower the quantization, lower the quality of the sound.

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

Data Subjected to deterioration Can be noise-immune without

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 8KHz: used for telephone

 11.025 KHz: Speech audio

 22.05 KHz: Low Grade Audio (WWW Audio, AM Radio)

 44.1 KHz: CD Quality audio

 Sample Resolution/Sample Size

 Each sample can be measured to a certain degree of accuracy.

 The accuracy is dependent on the number of bits used to

represent the amplitude (sample resolution).

IT 516 (Multimedia Systems) Compiled by Asefa M. (MSc.)

 8 Bit Value (0-255)

 16 Bit Value (Integer) (0-65535)

 The amount of memory required to store t seconds long

• is the sampling Frequency

 For 8 bit resolution, stereo recording

 For 16 bit resolution, mono recording

 For 16 bit resolution, stereo recording