Nice Picture. Pity About The Sound!

The other day, I decided to buy a new TV set. Although there’s nothing essentially wrong with the one I was going to replace (it’s only about five years old), it doesn’t connect to the internet, doesn’t have a coffee-making gizmo and won’t let me switch it on remotely (via the internet) so it is up and running when I get home with less than 30 seconds to spare before the program I have rushed back to see begins!

So I visited a few stores to see what the current state of the market was. Apart from being mesmerised and nearly going into a brainwashing trance by trying to watch 20 televisions simultaneously impinging on all angles of my visual field, I was surprised that not once was sound mentioned by the salespeople…until I asked about it. This was then accompanied by “You actually want to listen to the TV sir?” in a tone that implied I was quite mad. It is, of course, impossible to judge anything about the sound in a store and, although I tried looking at the audio specifications, they didn’t tell me anything worthwhile. Thus, I settled for the one I liked the look of best.

It is interesting that that the word “television” is all about the vision bit and not a mention of the sound. The same is true for “moving pictures,” i.e., “movies,” and not a hint of the auditory accompaniment. Nonetheless, as someone quite famous once said, “Sound: It’s 50% of the movie!”

The problem with being an audio guy is that I like to know what I am listening to. After a BBC program producer recently blamed the tuning of viewers’ (listeners’?) TV sets for the barrage of complaints about the poor dialog intelligibility of a particular program, I began to wonder about the audio performance of a typical TV set. The manufacturers’ budget for the audio section of a TV has always been disproportionately small, and since the introduction of LCD/LED flatscreens, there is also even less available real estate for the loudspeakers.

Therefore, not only do the loudspeakers have to be tiny, but to add insult to injury, they seldom point at you, but rather downward under the TV or even backward to the wall behind the set. Now, this immediately violates Mapp’s first law of sound system design and speech intelligibility that, simplified, basically states, “Point the speakers at the listener.” (Actually that’s exactly what it says in the full version but, for scientific purposes, I usually throw in some dBs, a transfer function and, if really struggling, a guest appearance by PI, as well).

Although many TV sets are wall mounted, countless others sit on top of (or even in) a piece of furniture or even a custom-made TV stand. The insensitive TV loudspeakers (in more ways than one) immediately invite the local boundaries to join in the fun and play a game of “audio comb filters.”

If you have ever wondered why you never see the frequency response of a TV system loudspeaker, then feast your eyes on the accompanying graphics showing the responses of a selection of TV sets I measured in five different rooms, and marvel at how we manage to decipher anything at all!

Figure 1. TV set with rear-facing loudspeakers; furniture mounted in room corner.
Figure 1. TV set with rear-facing loudspeakers; furniture mounted in room corner.

Figure 1 shows a high-resolution plot of a TV mounted near the corner of an acoustically benign living room. The peak in the response, that subjectively causes a “muddy boom” and muffles the intelligibility, is centered around 200Hz. This is followed by a series of comb filters and a sharp, premature roll off at 4kHz. Figure 2 shows another set, this time with the speakers mounted at the bottom of the screen firing downward. Although the response is much better extended, reaching out to 10kHz, the comb filters and associated peak at 200Hz produce significant coloration of the sound.


Figure 3 shows the 1/3 octave response of a TV set (stand mounted) with three different user settings applied. Although the boundary control and speech setting significantly reduce the mid-frequency, muffling boom, the peak at 2kHz is not at all pleasant.

Figure 3. TV loudspeaker responses for various user settings.
Figure 3. TV loudspeaker responses for various user settings.

Figure 4 shows the response of a wall-mounted set and corresponding “suck out” (phase cancelation dip) at 250Hz. The speech enhancement control provides a 3dB to 4dB boost over the consonant region, but this tends to make the already “thin” sound even thinner, though useful for some listeners/programs. A variation on the high-frequency enhancement can be seen in Figure 5 where, in this set, it is accompanied by a complimentary 8dB cut at 200Hz. Apart from the now-familiar 200Hz peak, the rest of this TV set’s response is fairly smooth and well extended.

Figure 4. Wall-mounted TV responses.
Figure 4. Wall-mounted TV responses.
Figure 5. Desktop-mounted TV responses, with speech enhancement filter.
Figure 5. Desktop-mounted TV responses, with speech enhancement filter.

So, if you have ever wondered why you don’t see published curves, now you know!

Playing the sound through auxiliary HiFi speakers or a sound bar often works wonders, but should we have to do this in order to enjoy reasonable sound at home? No wonder movie theaters are doing so well. After all, the sound is so much better there, isn’t it? Well, you will have to wait until next month to find out about that. In the meantime, where did I put the remote control for the coffee maker?

Previous ArticleNext Article

Send this to friend