Bringing Yourself Up to Speed with AAC, MP3, and Digital Audio - What Determines Audio Quality?
(Page 3 of 16 )
Having high audio quality means having enough data for your audio equipment to deliver playback that sounds not only recognizable from the original but also good. That’s pretty obvious once you think about it, but to get high-quality audio playback, you need several different things, including:
- A high-quality audio signal—for example, an uncompressed audio file (such as a CD track, a WAV file, or an AIFF file) or a compressed audio file (such as an AAC file, an MP3 file, or an Ogg Vorbis file).
- A high-quality playback device, such as your iPod.
- Good speakers or earphones. With speakers, you also need an amplifier, either built into the speakers or separate.
- Ears in tolerable condition. If you’ve lost hearing in some parts of the audio spectrum, you may need your audio equipment to compensate in order to make the music sound good to you. Even if you have perfect hearing, you still need to get music of a quality you like.
The first two items—the audio signal and the playback device—aren’t subjective. Most people can agree whether an audio signal is high-quality or not, because audio quality is measured by the amount of data conveyed rather than by whether you like the audio or not, and you can objectively compare the playback quality of different audio devices.
But the last two items—the speakers or headphones, and your ears—are highly subjective. If you like Limp Bizkit, you probably won’t get good mileage out of a system designed to deliver quality Liszt. If you prefer Beethoven, you’ll need speakers that can deliver the subtleties and the thunder of classical music rather than speakers designed to crank out country and western. If acoustic folk music is what brings light into your life, speakers or headphones designed to deliver stomach-churning bass and explosions in action games won’t do you much good.
Anyway, you get the idea, so enough of that for now. We’ll consider all these variables later in this chapter.
What Does CD-Quality Audio Mean? As you probably know, the human ear is an imperfect audio instrument, capable of detecting (and conveying to the brain) a limited set of audio frequencies that spans the range of experiences that humans evolved (or were designed, depending on your point of view) to know about.
Sound frequency is measured in hertz (Hz) and kilohertz (KHz): 1 Hz is one cycle per second, and 1 KHz is a thousand cycles per second. The typical human ear (let’s assume it’s yours, even if you’ve been listening to loud music as much as I have) can detect sound waves from about 20 Hz to 20 KHz. Not surprisingly, most music made by and for humans concentrates on these frequencies, with some (usually unnoticed) overlap at each end. Low-frequency sounds below 20 Hz are called infrasound, and high-frequency sounds above 20 KHz are called ultrasound. (The ultrasound devices used for physical therapy emit very high-frequency waves to subtly rearrange your tissues. If you hear these waves, you’re in trouble.)
NOTE: Animals’ hearing overlaps human hearing at both ends of the scale. Typically, dogs’ hearing ranges from 50 Hz to about 45 KHz, which is why they can hear dog whistles that most people can’t. Bats’ ears can hear up to 120 KHz, which is why they can navigate without bothering humans or dogs. Dolphins can hear up to 200 KHz, so bats might be able to annoy them in certain limited circumstances. Elephants can hear infrasound frequencies down to about 5 Hz, so bats don’t bother them but earth tremors may.
CD-quality audio is digital audio that includes the full range of frequencies that humans can hear (again, with some overlap at the high and low ends for the gifted and to annoy susceptible animals). The digital audio is created by using a process called sampling—taking snapshots of the audio stream to determine how a particular moment sounds.
CD-quality audio samples audio 44,100 times per second (a sampling rate of 44.1 KHz) to provide coverage with no gaps. Each sample contains 16 bits (2 bytes) of data, which is enough information to convey the full range of frequencies. There are two tracks (for stereo), doubling the amount of data. The data on audio CDs is stored in pulse code modulation (PCM), a standard format for uncompressed audio.
CD-quality audio consumes around 10MB (megabytes) of storage space per minute of audio—a huge amount of data even in these days of 250+GB hard drives. So you can see why you need compression to make a large amount of music fit on a device like your iPod.
How Compressed Audio Works (in Brief) Digital audio data tends to be difficult to compress using lossless compression because it contains so much rapidly changing information. So most methods of compressing digital audio data tend to use lossy compression, discarding the least important parts of the audio while trying to keep everything vital for the sound.
As you’d imagine, the first part of this reduction is getting rid of data on any frequencies that fall outside the hearing range of most humans. There’s no point in including any frequency that the vast majority of listeners won’t even be able to hear. (Ultrasound and infrasound fall by the wayside here.) But most audio codecs (coder/decoders) get more subtle than that. Besides taking into account the limitations of the human ear, the codecs use psychoacoustics, the science of how the human brain processes sound, to select which data to keep and which to discard. As a crude example, when one part of the sound is masked by another part of the sound, the encoder discards the masked part, because you wouldn’t hear it.
Did you Know? Infrasound Can Affect You Even When You Can't Hear It Even if your ears can’t hear infrasound, it may affect you. In summer 2003, scientists in Britain conducted an experiment to gauge the effects of infrasound on the listener. The scientists added infrasound bass lines to modern music works being performed and asked the audience to fill in questionnaires that asked how certain passages of the music affected them, using double-blind techniques to make the results as valid as possible. The infrasound passages produced unsettling effects such as increased heart rate, feelings of anxiety, shivers on the skin, and fluttering in the stomach. In some participants, the infrasound even evoked sharp memories of emotional losses. Pipes in some church and cathedral organs can produce infrasound frequencies down to around 16 Hz, which may account for some of the feelings some people experience in such places of worship. It may also explain why elephants tend to avoid churches. |
How much data the encoder keeps depends on a setting called the bitrate. Almost all encoders let you choose a wide range of bitrates. In addition, most MP3 encoders can encode either at a constant bitrate (CBR) or a variable bitrate (VBR). The pros and cons of CBR and VBR are discussed in the section “Choose Between CBR and VBR for MP3,” later in this chapter.
 This is chapter three of How to Do Everything with Your iPod & iPod Mini, by Guy Hart-Davis (McGraw-Hill/Osborne, ISBN 0072254521, 2004). Check it out at your favorite bookstore today.
Buy this book now. |
Next: What Is AAC? Should You Use It? >>
More Software Articles
More By McGraw-Hill/Osborne
| Recommended by Dev Hardware |
|---|
|
/ 69 
