Scribd
Upload
49 views26 pages

RT Lecture 5 Slides

1) This document discusses the differences between analogue and digital audio, how audio is converted between the two domains, and some key concepts like sample rate, bit depth, and compression formats. 2) It explains that analogue audio is converted to digital by sampling the amplitude at regular intervals, and that digital audio can be reconstructed through a DAC. 3) Key factors like sample rate, bit depth, lossy vs. lossless compression are defined, and how they impact aspects like frequency response and dynamic range.

Uploaded by

james k
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
49 views26 pages

RT Lecture 5 Slides

1) This document discusses the differences between analogue and digital audio, how audio is converted between the two domains, and some key concepts like sample rate, bit depth, and compression formats. 2) It explains that analogue audio is converted to digital by sampling the amplitude at regular intervals, and that digital audio can be reconstructed through a DAC. 3) Key factors like sample rate, bit depth, lossy vs. lossless compression are defined, and how they impact aspects like frequency response and dynamic range.

Uploaded by

james k
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 26

This subject is abstract and convoluted, but the key points to remember are only a

few and the applications are practical and simple.


You need this knowledge to understand:

1) What the difference is between analogue and digital.

2) How a signal is digitised and reconstructed.

And in practical terms:

1) What you get when you choose a specific sample rate

2) What you get when you choose a bit depth

3) What the application Is of different audio formats


Sound in an electrical signal is an analogue representation of
acoustic sound. Just like in acoustic sound, an electrical signal
carrying sound is continuous: it flows in an infinite number of
points.

An electrical signal is converted into digital by taking periodic


samples of the wave’s level at a predefined rate. This breaks up the
wave making it discontinuous generating enough values to
resemble the wave accurately.
• A digital recording is an accurate
representation of the sound wave in an
electrical signal.

1. This method produces an insignificant


amount of noise.
2. Storage requires minimal space.
3. Duplication of digital sound is 100%
accurate.
4. Digital sound can be processed to a high
degree of detail.
5. Audio can be edited non-destructively.
Here is the whole process:

Recording

Playback
C.
D.
A.

The ADC takes periodic samples of the wave’s amplitude (voltage


level changes) at a predefined rate (number of samples per second).

Analogue Wave ADC Sampling Digital Wave


e
i tud
pl
Am

Each sample registers a fixed voltage value(the wave’s


amplitude level when the sample is taken) which is stored
as numbers in bits.

Higher voltage

Lower voltage
The wave below is a zoom in to sample level with stem plots showing each sample.
The wave below is a zoom in to sample level showing
the sample steps of a wave in the digital domain.

This and the previous slides are the same thing, but this slide shows the most common
representation of a digital wave in digital audio software.
e
i tud
pl
Am Higher amplitude level uses more bits: the higher the
amplitude the higher the number in bits.

Higher number value

Lower number value


ate
R
The predefined sampling rate - the number of samples per second that
the DAC takes - determines how small can be a recorded wave (see next
slide).
Nyquist’s Sampling Theorem

The number of samples required to represent a wave is defined by the Nyquist


Shannon’s sampling theorem:

In digital audio, a recording must be band-limited and the sample rate


must at least double the desired frequency bandwidth.

So if the signal has a 10kHz component, the sample rate needs to be 20kHz or more for a
faithful reproduction.
The resulting frequency of sampling-rate / 2 is named The Nyquist Frequency

For 44,100 Hz sample rate, the Nyquist frequency is 22,050 Hz


For 48,000 Hz sample rate, the Nyquist frequency is 24,000 Hz
For 96,000 Hz sample rate, the Nyquist frequency is 48,000 Hz
Aliasing

The purpose of the Nyquist frequency is to prevent aliasing:

Aliasing is the distortion or artifact in digital audio consequence of a


signal above half of the sample rate.

To prevent aliasing, a high cut filter is used to eliminate frequencies above the
Nyquist frequency.
Aliasing example at 44.100Hz sample rate

Filter used to prevent aliasing

Aliasin
g

Sound Ultrasound
.C.
A
D.
The digital to analogue converter (DAC)coverts the wave represented in digital data
to an electrical current, by:
1) generating periodic fixed electrical voltages at a predefined rate (sample rate);
2) filtering the digital artifacts, restoring the regular wave.

DAC
Conversion

Digital Wave Restored Analogue Wave


The illustration below shows how the samples get converted into a
continuous wave by the DAC.
Sample rate sets the available frequency bandwidth. As in Nyquist theorem:
bandwidth = Sample Rate /2. The higher the sample rate the broader the bandwidth.

Bit depth, also know as resolution, sets the available dynamic range. The higher the
bit depth the higher the volume dynamic range. Each bit adds 6dB [approximately]
of volume, so for example 16 bit = 96dB of dynamic range.
•
•
•

The most common sample rate is 44,100 Hz, used in Compact Disk, YouTube,
Spotify, etc. In video and film, the standards are 48,000 Hz or 96,000 Hz.
96,000 Hz is also used in the professional audio’s production process, and for some
so-called ‘high definition’ audio products sold online.
•
•
• (described later on)

Bit depth 16bit, 24 bit and 32 bit are of standard use. C.D. is 16 bit and a common
format for finished products and 32 bit for mp3/AAC. Music is usually recorded in
24 bit resolution.
Lossy:
These formats are encoded using an algorithm that reduces file size by eliminating
elements in the audio that are considered not noticeable. They are named Lossy
because of the integrity of the file degrades in the process.

Lossless:
These formats keep the integrity of the audio data.
Compressed Audio Format

Loosy Formats
• mp3
• iTunes type

MPEG-1 or MPEG-2 Audio Layer III: MP3


• Bit rate: 32, 40, 48, 56, 64, 80, 96, 112, 128, 160, 192, 224, 256 and 320 kbit/s

Advanced Audio Coding: AAC


• Same bit rate
• iTunes format extension m4a
Uncompressed Audio Format

Lossless Formats
•Wav
•Aiff

Sample Rates
•44,100 Hz
•48,000 Hz
•88,200 Hz
•96,000 Hz
•176,400 Hz
•192,000 Hz
•384,000 Hz

Bit Depth
•16 bit
•24 bit
•32 bit

You might also like