In the digital era, most modern multimedia devices are only processing digital audio. Even in the case of mobile phones, there is the conversion of analog audio input required into digital output before transmission.
But if you prefer to watch a video instead, click here:
What is Digital Audio?
Digital audio is the imitation and transmission of sound in a digital format. In other words, digital audio is a technology used to generate, manipulate, reproduce, record and store sound using sound waves encoded in digital format(s). The set-up includes any sound files stored on a computer.
Likewise, digital audio is composed of discrete points representing the amplitude of the waveform rather than a continuous sinusoidal wave. These points are the sequence of separate samples taken from an analog audio waveform. Hence, the more samples in use, the higher the representation will be.
Before proceeding any further, let’s first learn about analog audio.
What is analog audio?
Analog audio is an audio wave recorded or used in its original form with the signals stored as a continuous signal—sinusoidal waves. These analog signals allow later playback for analog videos and analog audio.
Phonautograph and phonograph were the first mechanical systems used for recording analog audio. Later came the waveform development of wire recording, tape recorders, cassette tapes, vinyl records, disc records.
Why Should Analog be Converted to Digital Audio
In this digital era, most of the audio recorded today is in the form of digital files. Therefore, most modern audio players are catering to digital audio formats. This change is leaving audio purists and past generations having sentimental value for the physical storages in problems.
Change, however, is inevitable. The two groups need to change and adapt if they are to avoid the challenges presented by analog recorded audio. Such problems include the:
- Distortion occurring to the sound on tapes due to repeated usage.
- Misplacement of the physical record in an unknown location leading to loss of data
One advantage of digital audio files is the easiness of editing. For example, it is possible to automate the generation of digital audio transcripts as the record plays live. You can also add and mix many digital effects to them to:
- Enhance certain frequencies
- Simulate reverberation
- Change the pitch
Moreover, sharing digital audio via online communication platforms has become easier than sending it as a physical copy. Finally, digital audio is accessible remotely due to the emergence of cloud storage.
So now the question remains, “How to go from analog to digital audio.”
Going from Analog to Digital
Analog audio systems work on the principles of human hearing. Sounds beginning in the air as physical waveforms, get transformed into an electrical representation via a transducer, such as a microphone, and then stored. Reversing the process will recreate back the sound.
Before converting the signals back into physical waveforms via a speaker, amplification comes first. The nature of the sounds might change, but its central, wave-like characteristics will usually remain unaltered throughout the whole process. However, due to the electronic circuits having inherent noise presence, analog audio signals might be susceptible to distortion and noise.
Likewise, digital audio production is similar to the creation of analog audio but with a slight difference. Rather than converting the analog audio into electromagnetic signals, the digital audio chain begins by converting the analog audio signals into electrical signals.
Here is a short process of the digital audio creation: conversion of analog to digital, storage, conversion of digital to analog, then transmission.
Coding of electrical signals is crucial in combating any possible errors occurring during storage or transmission. Channel coding aids in the ability of the digital systems to replay the analog systems upon request. One channel code example includes an audio compact disc (CD) having 8-14 Bit modulation.
When converting an analog audio source into digital audio, the digital encoding shall require:
- sampling frequency
- The number of bits used in taking samples.
In the sampling frequency, devices will take thousands of samples per second to ensure the successful replication of the waveform. Each sample shall represent the intensity of the waveform in a specific instance.
As for the number of bits, 16 bits is the standard sampling parameter used; taken over a spectrum of 44.1 Kilo Hertz (kHz).
Therefore, the question remains, “What are the main factors Affecting Digital Audio Quality?”
Main factors Affecting Digital Audio Quality
The quality of digital audio mostly depends on two main factors:
- Bit depth or sample format
- Sample rate or number of bits
Increasing the number of bits in each sample will subsequently improve the recording quality. However, increasing the sample rate will also increase the amount of space occupied by audio files on a computer or portable storage device.
1. Sample Rates
A sample rate is the value representation of the waveform in terms of the number of samples captured per second. We measure the rates in hertz (Hz), or cycles per second. Higher sample rates can support the representation of higher audio frequencies.
It is possible to reconstruct the exact waveform from the digital samples. The sample rate needs to be more than double the highest quality audio currently present. Digital samples cannot correctly represent frequencies which are more than half the sample rate. “Half the sample rate” is the representation of an upper limit called Nyquist frequency.
When correctly representing the analog waveform digitally, it should be below this limit. Representing analog frequencies at this limit or above will be incorrect and would lead to a kind of distortion called aliasing. Before converting to digital, make sure you remove these frequencies if they are present in the original audio.
Usage of Sample Rates
The human ear is quite remarkable. It is sensitive to sound patterns having frequency ranges of approximately 20-20,000 Hz. If sounds are outside that range, they will be inaudible. Thus, 40,000 Hz is the absolute minimum sample rate essential in producing a full range of audible sounds. We use higher frequencies called oversampling to allow acceptable filtering and to avoid aliasing artifacts to the Nyquist frequency.
Telephones can only transmit frequencies ranging from 200 Hz to 4000 Hz. But for human speech, it is still intelligible even if we eliminate frequencies above 4000 Hz. Speech quality for audio recording commonly has 8,000 Hz as its sample rate.
Computer cards are limited to 48 kHz or 96 kHz, sometimes reaching to 192 kHz for higher performance ones. However, the most common sample rate in many cards is 44.1 kHz. Audio CDs also use this rate.
2. Sample formats/ Bit Depth
We measure bit depth by the number of computer bits representing each sample. Therefore, the more precise the representation of each sample is, the more bits are in usage. Increasing the number of bits tends to upsurge the maximum dynamic range of an audio recording. So, what is a dynamic range?
Dynamic range is the difference between the softest and the loudest sounds represented. We measure the difference in decibels (dB). For the human ear, 90 dB is the least dynamic range it can perceive. Hence, it is advisable to record digital audio with a dynamic range of more than 90 dB. This dynamic range allows for softer sounds to be amplified for maximum fidelity.
It is crucial to note that recording of low-level signals isn’t able to use all the available bit depth. The loss of resolution isn’t merely re-capturable by normalizing or raising the overall level of a digital waveform.
The most common sample formats and their respective dynamic range include:
- 8-bit integer: 48 dB
- 16-bit integer: 96 dB
- 24-bit integer: 145 dB
- 32-bit floating-point: near-infinite dB
Practical limitations of the dynamic range occur due to the capabilities of the input and output converters hardware. For example, the practical limit for a 16-bit is more like 90 dB. Most computer audio files and audio CDs use 16-bit integers.
What are the Types of Digital Audio Formats Available?
- AIFF for Apple computers and WAV for Microsoft stores audio without compression. These are not popular with casual listeners who prefer compression to save on disk space.
- Lossy formats are for compressed audio such as the mp3, FLAC or ALAC. Lossy means the removal of several parts of a file. Removing these parts should have the least effect on sound quality perception.
To casual listeners, the difference between the two might be not clear. They might need decent speakers or a pair of headphones to appreciate the difference. But to a sound engineer, they are quick to note it. Thus, audio engineers would prefer uncompressed files.
Regardless of the type, storage of the samples is usually in binary form. They are then merged into a single data file and formatted correctly for usage on a digital player.
How to Calculate the size of Audio Files
The following information might not be that vital because the advancement of technology has automated the process already. However, it is good to have an understanding of how the process works.
Determining the size of an uncompressed audio file will require the multiplication of:
- Sample format bit rate, e.g. 16-bit
- Sample rate, e.g. 44.1 kHz
- Number of channels, e.g. 2 for stereo, and
- Number of seconds
For a full 74-minute stereo audio CD, its take-up storage is 6 billion bits, which is less than 800 megabytes (MB).
What’s Next for Digital Audio
Digital audio gradually replaced analog audio technology in the 1990s and 2000s due to its advancement in the 1970s. The trend gave rise to audio standard technologies which include:
- HD radio
- Digital audio broadcasting (DAB)
- In-band on-channel (IBOC)
- Digital Radio Mondiale (DRM)
It also gave rise to the then-new audio-specific storage technologies such as:
- Digital Audio Tape
- Compact Disc (CD)
- Digital Compact Cassette (DCC)
By the end of this decade, expect the complete elimination of the audio-specific storage technologies. Files will only be stored as soft copies and played back on computers, smartphones, or in a cloud-based platform.