Learning Objectives: (WAGBA)
Understand how analog sound is converted into a digital signal through sampling.
Define and explain sampling rate, bit depth, channels, and bit rate.
Identify the hardware involved (ADC/DAC).
Calculate the file size of a sound file.
Sound in the real world is analogue. Think of it as a smooth, continuous wave, like a ripple in a pond. It has infinite detail.
Computers, however, are digital. They only understand discrete, specific values—0s and 1s. So, how do we convert the smooth analog wave into blocky digital data?
We take samples!
Sampling is the process of measuring the amplitude (height) of the sound wave at regular, fixed intervals. It's like taking a series of "snapshots" of the wave over time.
Imagine tracing the wave, but you can only put a dot down once every second. The collection of these dots is your digital representation of the wave. The more dots you place, the closer you get to the original shape!
Watch this very formative and well-explained video from Sam Loose, one of the dominant Audio engineerss in his own field at Izotope.
How good our digital sound is depends on two key factors: Sampling Rate and Bit Depth.
1. Sampling Rate (or Sampling Frequency)
This is how often we take a sample. It's measured in Hertz (Hz). A common sampling rate for music is 44,100 Hz, which means we take 44,100 "snapshots" of the sound wave every single second!
Low Sampling Rate: The digital version is a rough, blocky copy of the original. It loses a lot of detail.
High Sampling Rate: The digital version is a much more accurate and faithful representation of the original sound wave.
2. Bit Depth (or Sample Resolution)
This determines how much information we can store for each sample. It's the number of bits (0s and 1s) used to represent the amplitude (loudness) of each snapshot.
More bits = more possible amplitude values = greater dynamic range (the difference between the quietest and loudest sounds).
Imagine describing a colour. With a low bit depth (e.g., 2 bits), you might only have 4 options (2²=4) like "black," "white," "grey," "dark grey." With a higher bit depth (e.g., 8 bits), you have 256 options (2⁸=256), allowing for many more shades and a more detailed description.
3. Channels
This refers to how many streams of audio are being recorded.
Mono (1 channel): A single audio stream. All sound comes from one direction.
Stereo (2 channels): Two separate audio streams, one for the left speaker and one for the right. This creates a sense of space and direction.
Convert 44,100 Hz to KHz.
The conversion process requires special hardware, usually found on a computer's sound card or motherboard.
So how does it actually happen?
An input device like a microphone captures the analogue sound wave and converts it into an analogue electrical signal.
The Analog-to-Digital Converter (ADC) receives this signal. This is the component that performs the sampling process to turn the analogue signal into a stream of binary data.
This digital data is then processed and stored by the computer.
To play the sound back, the process is reversed. The Digital-to-Analog Converter (DAC) reads the binary data and converts it back into an analogue electrical signal.
This signal is sent to an output device like speakers or headphones, which vibrate to recreate the sound waves for your ears to hear.
Drag the components onto the correct stage of the sound journey.
DAC
Speakers
Headphones
ADC
Microphone
Recording
Playback
Now we can put everything together to figure out how much space a sound file takes up!
First, we calculate the Bit Rate: the number of bits of data that are processed every second.
Formula: Bit Rate (bps) = Sampling Rate (Hz) x Bit Depth x Number of Channels
Next, we use the bit rate to find the total file size.
Formula: File Size (bits) = Bit Rate (bps) x Time (seconds)
To convert this to bytes, remember: 8 bits = 1 byte.
Example:
Let's try an example. Calculate the file size in megabytes (MB) for a 3-minute (180 seconds) stereo (2 channels) audio track with a sampling rate of 44,100 Hz and a bit depth of 16. (Assume 1 MB = 1,000,000 bytes).
Bit Rate = 44,100 x 16 x 2 = 1,411,200 bps
Total Bits = 1,411,200 x 180 = 254,016,000 bits
Total Bytes = 254,016,000 / 8 = 31,752,000 bytes
Total MB = 31,752,000 / 1,000,000 = 31.75 MB
You are recording a 1-minute (60 seconds) podcast. It is a mono recording with a sampling rate of 22,050 Hz and a bit depth of 8. What is the file size?
rate
microphone
amplitude
digital
depth
larger
sampling
analogue
KiloHertz (KHz)
The number of times a sound wave repeats per second is called its 'amplitude.'
Sound is converted to an audio file by taking ‘samples’ of its amplitude at different points in time.
Speakers produce sound by vibrating a diaphragm.
The number of possible values for amplitude of a sound is a quality we call ‘bit rate’.
Metadata contains information about the sound file as a whole.
Each audio channel has its own samples.
The height of a sound wave is called its 'amplitude.'
Sound is converted to an audio file by taking ‘samples’ of its amplitude at different points in time.
Increasing the number of samples you take per second will decrease the size of an audio file.
Match the key terms with the statements.
| Draggable item | arrow_right_alt | Corresponding Item |
|---|---|---|
Bit depth | arrow_right_alt | Taking amplitude values of a sound wave |
Sampling | arrow_right_alt | Separate audio track with its own sampling |
Metadata | arrow_right_alt | Number of times a sound wave repeats per second |
Sampling rate | arrow_right_alt | Height of the sound wave |
Frequency | arrow_right_alt | Range of possible values for amplitude when sampling |
Hertz | arrow_right_alt | Bits of data per second of audio |
Bit rate | arrow_right_alt | Information about the sound file, included with the sound file |
Amplitude | arrow_right_alt | Unit of frequency |
Channel | arrow_right_alt | Number of samples taken per second |
Describe the function of three pieces of hardware involved in the capture or production of digital audio.
Order the following steps for the 'sampling' process.
Each of them is taken in as a single ‘sample’.
Amplitude values are taken at different points in time.
These values are converted into binary through ADC.
Thousands of samples are taken per second.
The binary values are stored as an audio file.
In your own words, explain to a friend why a song streamed on Spotify (high quality) has a much larger file size than a voice message you send on your phone (low quality). Use at least two of the key terms we learned today (e.g., sampling rate, bit depth, channels).
What does sampling rate measure in sound processing?
How does higher bit depth affect sound quality?
What does a single channel in audio mean?
What is a common sampling rate for music?
What does an input device like a microphone do?
What does the Analog-to-Digital Converter (ADC) do?
What is the role of the Digital-to-Analog Converter (DAC)?