Audio Performance

Last Updated: 23-Jul-2008

Along with RF performance,  one of the biggest differences between cell phone models is their audio quality. Unlike RF however, audio is much harder to quantify. For starters, there is absolutely no industry-accepted standard for how a cell phone should sound, and few in the industry seem to take it very seriously. Some companies seem to have a total disregard for how their phones actually sound, while others seem to try very hard to produce phones that have excellent aural properties.

Audio quality can actually be broken down into two very distinct facets. The first is the phone's ability to reproduce the audio without adding distortion or unwanted noise. I call that Sound Reproduction. The second is the Tonal Balance of the resulting audio. Note that a phone's ability to reproduce a true rendition of the audio will vary from technology to technology, simply because GSM, CDMA, IS-136, and iDEN use different CODECs. A CODEC in the cell phone world is always a compromise between quality and using the fewest possible bits. It is rarely possible to compare phones that use different technologies, but we can compare phones that use the same technology (GSM vs GSM, CDMA vs CDMA, etc).

Tonal balance is probably the most subjective of all, since different people have different ideas about what constitutes good tonal balance. My opinions on this matter are just as subjective as the next guy's, but I do pride myself on being an audiophile, and so I believe that I have a fairly good handle of what's balanced and what's not. However, that doesn't mean that my opinion is necessary in keeping with what you personally want from a phone.

When reading my reviews, your best best is to use my tonal balance comments as a guideline, and compare them to what I've said about other phones (some of which you may already be familiar with). As for the phone's ability to reproduce the audio however, that's much less subjective, and I do believe that my ability to detect distortion and other unwanted audio maladies is reasonably well honed.

Once a flaw is recognized, the problem isn't whether it exists, but to what extent it really matters. This will depend upon your own personal threshold of tolerance when it comes to such matters.

Much of what we accept as good audio quality from a phone is merely relative to what we were used to before. For example, when I first switched from the analog world to the digital world I bought a Nokia 2190 on Fido. I thought that phone sounded incredibly good, since it was better than what I'd been accustomed to up to that point. I still have that 2190 sitting around, and sometimes I charge the battery and pop my SIM into it just to hear what it sounds like. If I were rating that phone today I would probably give it such low marks for audio reproduction that I'd be recommending that no one wasted their money on it.

So even on less subjective matters as audio reproduction, background noise, etc, it still comes down to your own personal likes, dislikes, and tolerance levels. As with tonal balance, it is best to read my comments simply as a guideline. If you're one of those people to whom all phones sound just fine, then consider yourself fortunate that audio quality isn't an issue for you. If audio quality is an important however, then I can only hope that my comments help you to make the right decision for you.

When it comes to testing audio quality of a phone, I try to subject the model to as many different types of voice as possible. For example, some phones might sound great on female voices, but muddy on male voices. I phone a variety of pre-recorded messages, and I carry on two-way conversations. As for outgoing audio, I do ask callers to give me their opinion of the sound, but I don't rely on that. Instead, I call my voicemail using the test phone, and then I compare the overall quality against a reference phone.

Outgoing audio tests are carried out under a variety of background conditions, ranging from quiet rooms, to noisy shopping malls, to noisy vehicles. I do this to see how well the phone in question copes with the background noises, and how clearly the user's voice is reproduced in each situation. Many CDMA phones for example are horrendous when background noises are quite loud, but they sound great in quiet environments.

Audio Glossary

Tonal Balance: This refers to the balance of sound levels across the full range of audio frequencies. If we say that a phone has good tonal balance, we mean that it doesn't enhance or suppress any particular set of frequencies. The audio is said to be "natural sounding".

Sound Reproduction: This is the ability of the phone to render a faithful copy of the caller's voice. Any distortion of the voice would result in a poor Sound Reproduction rating.

Peaky: This is a sign of poor tonal balance. It means that one particular set of frequencies has been greatly enhanced such that elements of speech that fall within that range sound far more pronounced.

Shrill: This is a specific type of peakiness, in which the sound contains too much high frequency energy. The most obvious facet of a shrill sound is that it often has a piercing quality to it that makes you pull the phone away from your ear. Never confuse this for high earpiece volume (though it may go hand-in-hand with that).

Tinny: This is when audio contains too much "high end" and too little "low end". A cheap AM radio with a small speaker would be an example of a tinny-sounding device.

Boomy: This is when audio contains too much "low end". If you were listening to music through a boomy sound system, the bass instruments (like drums and bass guitars) would sound overly loud, and they would produce odd sound effects reminiscent of a huge drum.

Muddy: This is quite similar to Boomy, but it lacks the "big drum" effect. It is usually rather indistinct sounding because of its lack of higher frequency components. Nuances of speech are much more difficult to discern.

Thin: This refers to sound that lacks many of the important frequencies necessary for voice reproduction. Even though the resulting sound might be perfectly understandable, you are aware that the voice is somehow "incomplete".

Distorted: This is when the audio is changed in some way. Distortion comes in many forms, but all of them alter the way the waveform is recreated.

Scratchy: This is a particular type of distortion in which a sort of scratching (or high-rate crackling) sound can be detected beneath the spoken words.

Coarse: This is when the voice reproduction seems to be less than smooth. It is hard to put into words, but it is best described as the audible equivalent of a grainy photograph.

Grainy: Same as Coarse.

Harsh: Usually the same as Grainy, but it can also encompasses other audio problems that make the voice quality less-than-natural.

Hiss: A background noise that doesn't directly affect the overall sound quality. Hiss is the result of random energy across the entire audio spectrum. It is generally referred to as "white noise". When the hiss is composed primarily of higher frequency components, it is referred to as "blue noise".

Crackling: This is another type of background noise that doesn't directly affect the overall sound quality. It is similar to hiss, but it contains random noises that make it sound a bit like an old phonograph record.

Rustling: A muted example of Crackling.

Sibilance: This is a condition in which "s" sounds become a bit scratchy and overly obvious. It's sort of like the way Sylvester the Cat sounds, but not quite so put-on, and more annoying.