Phone audio quality sucks

Seth Godin says what we’ve all been thinking: “Am I the only person who wants a Hi Def telephone? A headset that sounds better than the handheld receiver”

It’s true. Twenty years ago, a hard-wired phone line sounded just fine. After a bumpy start with the 40-50 Mhz bands, cordless phones started going places at 900 MHz in the 1990s, too.

Then our local EM fields got too busy, supposedly, for this wavelength to work well. Cordless phones moved to higher frequencies, which meant less interference–but the quality never seemed there, notwithstanding the cascade of technobabble printed on the boxes. I’ve owned at least a dozen 2.4 GHz and 5.8GHz models, but none have ever matched the audio quality of a Panasonic 900MHz model I still own to this day. In a house packed with strange EM fields, it sounds just fine, too.

Thence to cellphones. I thought Sprint was bad, but then I tried to make calls on my wife’s iPhone. AT&T voice quality is just abysmal: it turns Apple’s amazing handset into a joke about the inverted priorities of futurism.

So what do we do with this stuttering, fading echo of the human voice? We pipe it through BlueTooth, just to make sure it sounds as bad as it possibly can.

About Rob Beschizza

Rob Beschizza is the Managing Editor of Boing Boing. He's @beschizza on Twitter and can be found on Facebook too. Email is dead, but you can try your luck at besc...@gmail.com
This entry was posted in Uncategorized. Bookmark the permalink.

25 Responses to Phone audio quality sucks

  1. codeman38 says:

    Yeah, as several people have pointed out, this is as much an issue of landlines, and the phone system in general, as of cordless and mobile phone bandwidth.

    Simply put, I’ve never, even on a wired phone, found the sound quality to be that great. Sure, some are better than others, but consonants have *always* sounded muffled to me over the phone due to that 4KHz cutoff. Combine that with a bit of auditory neurological weirdness on my part, and I have to strain quite often to make out what people are saying over the phone.

    And interestingly enough, this *isn’t* an issue with Skype. There are people who I have great difficulty understanding even on a land line, but who I have no trouble understanding over the much higher frequency range of Skype.

  2. Mike says:

    I have long fantasized of what you term a high-def phone. I thought of it instead using the old “CD quality” term, dreaming of double the landline 64-kb/s channel bandwidth, true full-duplex and cellphones with decent speakers. With 3G services there’s no reason this couldn’t be possible… but all that bandwidth is getting sucked up with other ‘value-added’ services.

    Streaming Nascar.

    Streaming… everything but a decent audio connection to the other human being at the other end of the wires (other end of the wireless?)

    But GSM audio compression expects what, a 19.2 kb/s channel? THREE TIMES LESS than a landline?

    You know, it’s probably this reason that AT&T put the hit out on GrandCentral / Google Voice. They didn’t want people to hear what the iPhone could really sound like instead of what people have gotten used to.

  3. Chris says:

    I use AT&T with an iPhone. before the iPhone I had a Sony-Ericsson W810 and before that a Motorola V600. Voice quality on the Moto was excellent, the Sony was pretty good, and the iPhone is just adequate. I think it’s the iphone, not the network.

  4. icky2000 says:

    @DEAN: Take a look at http://en.wikipedia.org/wiki/.tel.

    @EVERYONE: I know nothing about cell phones and bluetooth and all of that but I know that phone companies (including VOIP providers) use a measurement called the MOS (Mean Opinion Score) to measure voice quality and that most voice systems calculate MOS values on the fly for calls and log the data for later analysis and troubleshooting. Especially in the world of VOIP, achieving acceptable MOS values absolutely requires Quality of Service network management to give voice packets priority over intermixed data packets. Without it, voice quality is dismal. I wonder if they do similar kinds of analysis on wireless networks?

  5. skeletoncityrepeater says:

    I think wireless phone audio quality is subject to so many variables, it will take a long time for it to get right. Think of all the different compression algorithms used for every phone. Think of all the low-quality transducers used in many peoples’ cheap phones (including my trusty old Nokia). Think of the compression used to efficiently transmit data over large distances, and over satellite, without lag or interruption. Then we have the uncompression and playback of those sounds. There are many layers of data transformation and compression, etc., that exist between the 2 phones in use. If you had a small closed network with nice modern phones like iPhones, things would probably sound great. When I have used Skype or any teleconferencing programs with decent microphones and conversion, it has sounded like I was in the room with the other people.

    Basically, a Hi-Def Telephone is probably out there, but will only make you sound a little better, and will only make the other person sound a little better.

    There is an advantage to low-quality, compressed audio, though. If you’ve noticed, audio on HDTV has a huge dynamic range – a lot of louds and softs. The commercials blare at highly compressed, and annoying, levels. This type of dynamic range wouldn’t be very effective except in a nice silent environment. Nice, thin, distorted voices cut through traffic noise, subway chatter, etc.

  6. skeletoncityrepeater says:

    In addition, isn’t the 900 MHz on the cordless phone is the bandwidth of the transmission from the handset to the terminal.. isn’t the rating on the wireless cell phone the quality of the transmission from the phone to the ‘cell sites’ distributed about the land? That figure doesn’t really have to do with the audio quality itself.. Chances are when we used to use home phones we were also usually calling our friends in a local network with fewer layers and shorter distances, and less noise. Those days are gone.

    I think the iPhone itself probably has decent to good audio quality but they can’t control what happens to the data after it leaves the phone..

  7. cans says:

    When choosing a cell phone, it’s important to look very carefully at call quality. There’s a reason the phone on my desk is an HTC Touch Pro and not an iPhone 3GS. My call quality easily rivals or exceeds that of a landline cordless phone. And that’s the point. It’s good for text, email, web and all of that—not as great as the iPhone for those uses but still good.

    Somewhat counter-intuitively, the rise of texting and mobile internet has changed what people look for when buying cell phones. Personally, when I’m looking to buy a cell phone, I’m looking to BUY A PHONE. Everything else comes second.

  8. dtweney says:

    Who cares? A phone is for texting.

  9. richard baguley says:

    This is a tough one, because whatever type of phone system you are using, your voice is getting squished, clipped, squeezed and otherwise messed around with in all sorts of unpleasant ways. And having the best phone in the world isn’t going to make any difference if the signal then gets squashed beyond all recognition on the network.

    Cell phones in particular squash the voice down to about 9.6kbps (although the codecs used vary, and there are several that go even lower than that if you are on a particularly busy cell site), and then there are all sorts of other issues that can lead to a crappy signal.

    Sure, a phone network could come up with a better quality level of service where it was compressed less, but unless the person on the other end was also using that, it would still sound lousy.

  10. chad sergerie says:

    The frequency given to a portable phone will not actually give its bandwidth. It’s usually the carrier frequency for the wirless transmission. In most cases, the higher the frequency, the more space needed between channels which allows for more bandwidth and higher quality modulation of the audio. Systems such as GSM are newer and are much better designed. They also allow for TDMA and FDMA to switch through many channels which can allow for abit higher of a bandwidth.

    The second reason why that old 900MHz telephone is better than the rest it that they where only allowed to be made int he early 90′s where i believe they split the bottom portion of the ISM band. These may have a much wider bandwidth than the newer cordless phones we see.

    Also, the reason why we may not see high definition phones is due to the limited bandwidth of the actual audio found on phone lines. I think you’re limited to about 3kHz of badwidth. This limits you to hear purely vocal audio. That is also why the music sucks when an opperator puts you on hold…

    Overall I think it’s due to the current phone line system which is stopping us from seeing better quality phones. If we want better phones we need to change everyones phones. I could see this hapenning sometime in the near future though…

  11. milovoo says:

    The iPhone audio seems decent enough when listening to music through the internal speaker, even a low-bitrate ringtone sounds quite good.

    Yet the sound quality is such that I generally skip talking and just send an email or text message when possible. I agree that it would be nice to have the sound quality of even the little StarTac I gave up somewhere around 2001.

  12. Art Vandelay says:

    Best cell phone call quality I ever had was on my old Motorola StarTac, which I would sometimes force into an analog connection. It got hot as hell, and the battery life suffered horribly, but the sound was much more like a cordless phone than a cell phone. Even the duplexing seemed to work right. Verizon eventually killed their analog service altogether, AND made me get rid of the StarTac because it wasn’t GPS-enabled.

    Lack of decent duplexing is the biggest problem with cell phones. Remember when both you and the other person could hear each other speak at the same time, instead of the walkie-talkie like effect we have now?

  13. Crispy Critter says:

    Regarding cordless landline phones, I’ve had great sound quality from a Uniden DECT system.

    In the US, it works in a 1.92-1.93 GHz band specifically dedicated for DECT systems, rather than sharing the 0.9, 2.4 or 5.8 GHz band that other cordless systems use. (The European version uses 1.88-1.90 GHz.)

    The DECT system replaced an older Uniden 5.8 GHz system that had rather poor range; with the new phone, I get a good signal anywhere in the house.

  14. Anonymous says:

    This would be a much more meaningful discussion if it were clear which codecs were being used.

    The EFR codec has been around since 1997 to cater for higher quality audio, and every phone I’ve had for the past 10 years has been able to use it.

    Do people just not know it exists or has everyone enabled it on their phones and still found it unsatisfactory?

  15. Rob Beschizza says:

    Absolutely, it’s the network. Some of the problems are inherent, because we all live behind trees and wet brick walls and so on. But some of it is just lack of infrastructure, codecs that squeeze every last bit, and inexplicable differences between carriers even in well-served locations.

  16. John Bourke says:

    The question is, how do you know that the iPhone has good sound quality? It could be crap, and you’d never know.

  17. dculberson says:

    It also has to do with the carrier’s unbelievable stinginess with bandwidth. It’s not enough that they have the voice clamped down to about 4 KHz, they also have to blank / mute the stream when they don’t sense audio above a certain level. That means when people pause, there’s a dead silence and a bit of a cut when they start speaking again. It makes hold music that has quiet passages completely bizarre.

    I was recently on hold with Sprint – over call quality issues, coincidentally enough – and their hold music sounded like some bizarre robotic jungle drummer being transmitted through tin cans and string. Because all the instrumentation was too quiet to trigger the audio threshold for transmission. I found it sadly appropriate that even the carrier can’t get their sounds across their own network.

    I don’t understand why, in an era of 3G networks, we’re stuck with 4KHz voice bandwidth. Go hog wild and give us 16KHz and we’d be blown away.

    Skeleton, it’s not just what you describe – land line calls from coast to coast in the analog days were still things of beauty compared to the digital crap we deal with now. Land lines are still better than cell phones nowadays – which makes no sense seeing as how they’re also digital past the local loop. They’ve just squeezed things too tightly, bandwidth speaking. Someone told them a human voice was intelligible at 4KHz so they said, “Great! Let’s do that!”

  18. JT Montreal says:

    Higher voice quality has been part of the GSM network for a long time: the AMR-WB codec. Problem: It’s not mandatory, so vendors and carriers don’t implement or enable it. There is no proper tandem-free agreement: even if YOUR carrier with YOUR handset can do wideband (50Hz-7kHz), as soon as you cross from one carrier’s network into another, you’re going over the POTS (plain old telephone system) with it’s 300Hz-3kHz bandwidth, due to the old design, even when it’s digital 64kbps 8k-sampled 8-bit alaw or ulaw PCM.

  19. Anonymous says:

    Thanks, Rob, but I’m still confused. Nothing to do with Zikman’s post I’m afraid :)

  20. zikman says:

    ^^^thanks I’ll get right on it!

  21. dssstrkl says:

    Call audio sounds just great on my iphone… when I use Skype!

  22. Rob Beschizza says:

    For the confused, Zikman was replying to some weird-as-hell spam.

  23. Dean Putney says:

    I want DNS for phone numbers too gawdemmit! I don’t want to remember ten numbers when I can’t even remember your whole name. Let me write an A record and get over with it.

  24. cheem says:

    The 900MHz open band is relatively narrow and so devices sharing the band tend to interfere with each other. Of course, most phones now don’t use the 900 MHz band. 900MHz also has a much larger range for a given power output due to an inverse square law (signal strength is inversely proportional to both the distance from the device and the frequency which the device is emitting at). They require proportionately larger antenna to work, though.

    The 2.4GHz band happens to contain the resonant frequency of water. This is more of a concern due to the microwave ovens than anything else, although trying to pick up a 2.4GHz signal through an aquarium is not likely to work. Since all 802.11b/g devices and bluetooth headsets and microwave ovens share this band, it’s no wonder analog signal quality is usually crappy here.

    As for the crappy quality you get over cellphones, it’s all a matter of bandwidth…

Leave a Reply

Your email address will not be published. Required fields are marked *

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

 

More BB

Boing Boing Video

Flickr Pool

Digg

Wikipedia

Advertise

Displays ads via FM Tech

RSS and Email

This work is licensed under a Creative Commons License permitting non-commercial sharing with attribution. Boing Boing is a trademark of Happy Mutants LLC in the United States and other countries.

FM Tech