If you are in your late 40s or older, then you likely grew up with a turntable or tape player in your home as a way of listening to store-bought music. Between 1982 and 1983, the compact disc entered the market and forever changed the way music was stored and transported. In this article, we are going to look at how digital audio works and dispel some of the myths around the conversion between the analog and digital domains.

What Is Digital Audio?

In the simplest of terms, a digital audio file is a representation of an analog signal using a series of digital words. In the digital domain, i.e., a computer, information can be stored as a 1 or a 0.

Computers can combine strings of 1s and 0s to represent characters in a text document, colors in a photograph, commands in a program or voltage levels in an audio file.

For decades, the standard for storing audio in the digital domain has been the Red Book Compact Disc Digital Audio (CD-DA) standard of a 44.1kHz sampling rate with a depth of 16 bits.

The sampling rate describes how often a voltage level is measured and stored. To capture the entire audible spectrum of sound, the Nyquist-Shannon sampling theorem states that the sampling rate needs to be at least twice as high as the highest frequency you want to record for it to be recreated with accuracy.

The second consideration in converting an analog signal to the digital domain is the need to store an adequate amount of resolution to properly represent the original signal. The Red Book standard uses a digital word length of 16 bits. This means that there is a string of 16 1’s and 0’s that can be used to represent 65,536 voltage levels. If you are converting the output of a microphone to digital, and the maximum voltage is 1 volt, then a resolution of 16 bits means that the resolution is 0.000015258789 volts. That’s a lot of detail.

Finally, the Red Book standard states that two channels of audio will be sampled simultaneously to create a stereo recording.

Some Quick Math on CD Quality Audio

For those interested, it’s easy to calculate the effective bitrate of a CD-quality audio file. Since we sample the audio signal at 44,100 times a second, and each sample has a voltage level represented by a 16-bit word, and we do this for two channels, 44,100 times 16 times two is 1,411,200, or 1.411 kilobits per second.

To calculate how much space it would take to store a song like “Bohemian Rhapsody” by Queen, you can simply multiply 1,411,200 by the number of seconds in the song (in this case, 355 seconds) for a total of 500,976,000 bits, or about 60 megabytes of data.

How Are Digital Audio Files Created?

A device called an analog-to-digital converter (ADC) is responsible for taking the analog signal and converting it into a digitally represented value. These devices are commonplace and are found connected to the microphone in your smartphone or the Bluetooth microphone in your car. They are incredibly compact and, relative to when they were first introduced, inexpensive.

ADC work in several ways, but we’ll describe the basics. Imagine, if you will, a series of comparator switches, each stacked one atop the next and referenced to an ever-increasing voltage. We’ll keep the example simple and say that we have eight switches, each of which is triggered at

0.125-volt increments. If we feed an analog signal into our comparator switch tree with a level of 0.3 volts, the bottom two switches will turn on, and we get the digital word 0010 (which is 2). If we increase the voltage to 0.8 volts, we trigger all but the last two switches and get the word 0110 (which is 6).

Counting in Digital

Counting in digital is easy, once you understand how it works. Each space in a digital word represents a value of 2 to the power of the location. So, the first space is 2 to the power of 0, which is 1. The second space is 2 to the power of 1, which is 2, the next space is 2 to the power of 2 which is 4, and so on.

To encode a value using this format, we simply assign a 1 or a 0 to each placeholder such that the sum values represented by the placeholders with a 1 represent the original value.

In our example above, we are using a very low resolution of 3 bits, which means we can show only eight different levels. This limited resolution, of course, introduces some error – known as quantization error. The math can get very complicated very quickly. Suffice it to say that in our example, our theoretical digitizer doesn’t know the difference between a voltage of 0.63 and 0.73 volts. This is a large error and would not work in an attempt to sample audio. Luckily, our 16-bit resolution gives us 65,536 levels from which to choose.

What About Those Crazy Stair-Step Graphs?

You have undoubtedly seen marketing images showing a comparison of CD-quality audio resolution versus high-resolution 96 kHz, 24-bit audio.

While the concept of having a higher sampling rate and more resolution is accurate, it doesn’t mean that the CD-quality audio signal suffers in any way.

To demonstrate this, we created two 20 kHz test tones in Adobe Audition. The first track has a 96 kHz sampling rate and a resolution of 24 bits.

As you can see, the waveform looks smooth and detailed and shows roughly five samples per cycle.

The second track is the same 20 kHz sine wave stored at a sample rate of 44.1 kHz and a resolution of 16 bits.

As you can see, there is no significant difference in the shape of the two waveforms. More importantly, they both look like sine waves and neither has any stepping in them.

Understanding Digital Audio

Storing audio signal in the digital domain offers distinct packaging and reliability benefits over analog storage media like vinyl records and magnetic tapes. Of course, digital files don’t degrade over time. Digital files are also impervious to playback speed issues. If your turntable or cassette deck is playing too slowly, the music won’t sound right.

In a future article, we’ll look at the file format options available for storing digital audio files. Until then, be sure to drop by your local specialist mobile enhancement retailer to see all the latest digital media-compatible source unit upgrades available for your car, truck or SUV.

This article is written and produced by the team at www.BestCarAudio.com. Reproduction or use of any kind is prohibited without the express written permission of 1sixty8 media.

We recently talked about how the conversion from analog to digital works and explained some of the basic terminology associated with digital audio files. In this article, we are going to look at how different digital audio compression algorithms work to reduce file size so that files can be shared more easily.

The Basics of Digital Audio

As we explained previously, sampling an analog audio waveform at CD-quality resolution requires 44,100 samples per second with a resolution of 16 bits for a pair of stereo channels. This results in a data stream that is 1,411 kilobits per second. For a one-minute long song, you’d need to store 84,672,000 bits of information. That’s about 50 megabytes for a five-minute song.

When we store audio in an uncompressed format, the information in the file doesn’t affect the size of the file. We could have a track containing a recording of a symphony orchestra, an audio test track or the last, last, last performance of the Rolling Stones. The size of the file will be the same if the track length is the same.

Lossless Audio Compression

Reducing file size has always been a concern when it comes to transmitting a file. Back when we had dial-up internet service, it would take hours to download a whole song in an uncompressed format. Acoustic modems had an optimal transfer rate of about 300 bits per second. High-speed analog modems reached a peak of 48 kilobits per second using data compression algorithms. To transfer high-quality audio, it would take about 30 seconds for every second of music. See the problem?

One of the most popular and well-known file compression methods is to zip a file. Zipping a file analyzes the content of the file and replaces repeated strings of data with a shortcut to identical information. When you unzip the file, you get the original back without any modification.

In terms of audio compression formats, the most popular lossless formats are Free Lossless Audio Codec (FLAC), Apple Lossless Audio Codec (ALAC) and the Monkey’s Audio (APE) format in a distant third place.

In terms of compression, converting our 70.69 megabyte audio file to FLAC results in a file size of 37.5 megabytes. This is a reduction of about 50 percent with no loss of sound quality, accuracy or detail.

Lossy Audio Compression

Whether you are trying to shrink an audio file or a photograph, one of the easiest ways to reduce the file size is to throw away some of the detail in the original file. For audio files, this often means limiting high-frequency information and reducing the detail of or eliminating low-level signals.

If we convert our original audio file to a 320 kbps MP3 file, the file shrinks to an amazing 16.0 megabytes. We do so by throwing out audio information that is difficult to hear. For example, if there is a loud guitar riff in one channel, the compression algorithm can dramatically reduce the detail of the relatively low-level audio information in the other channel without much change in the perceived quality of the playback. This is called perceptual audio encoding because the algorithm specifically affects information that is more difficult to hear (or perceive).

At a compression rate of 320 kbps, most listeners can’t tell the difference between the original file and the compressed version. As the compression increases, the differences become much more apparent. We start to lose high-frequency information and detail.

If we want to shrink the file further, we can convert it to a 128 kbps MP3 file. The benefit of extreme data compression is that our audio file now has a size of 6.59 megabytes. At this size, the song can be attached to an email without much concern for bandwidth or download time on a modern broadband internet connection.

The most popular lossy file compression formats are MP3 (formerly MPEG-1 Audio Layer III or MPEG-2 Audio Layer III), Adaptive Transform Acoustic Coding (ATRAC), Advanced Audio Coding (AAC) and Windows Media Audio (WMA).

Does File Size Matter?

As mentioned, audio compression algorithms were created to allow us to transmit audio over limited-bandwidth connections. Today, data storage is incredibly inexpensive. You can buy a 128 GB USB memory stick for less than $25. You can store about 400 hours of high-quality FLAC audio on a 128 GB memory stick. At the same time, internet bandwidth speed is at an all-time high. Most smartphones with LTE can download data at 150 mpbs. That’s faster than most people’s high-speed internet at home. Downloading a 37 megabyte file over a connection like that takes about five seconds.

Unless you are bandwidth-limited, you may as well download your music in at least CD-quality FLAC or an uncompressed WAV format. That way, you get the best sound quality possible from your audio system. If you have questions about how many tracks you can store on a memory stick or what digital media file formats are compatible with your car radio, visit your local mobile enhancement retailer. They should be able to answer any questions you may have.

This article is written and produced by the team at www.BestCarAudio.com. Reproduction or use of any kind is prohibited without the express written permission of 1sixty8 media.