Why tolerate bad phone audio?

jes

jes

MyBroadband Alumnus
Joined
Nov 11, 2009
Messages
11,992
Reaction score
123
Why tolerate bad phone audio?

Despite all the technological advances in the communications environment, the quality of voice communication has not improved much from the early-day carbon microphone.
 
My guess is, bandwidth. If the fall back of all cell phones are GSM, which is the minimum. Then we should be looking at upping the minimum. How do you tell the poor to throw away their old crappy phones and upgrade to more than what they would earn in a year cell phone? :wtf:
 
chopsuey said:
I normally like to pick a fight with people who talk like this ... but don't have the energy today :p
Click to expand...
Funny how one is at work, things are very quite, not a lot of work to do and then I am ten times more tired! I start to understand how the public sector is working!
 
chubster said:
My guess is, bandwidth.
Click to expand...

Yip

* If we are only using 20% of the freq range as the article states, would people be prepared to pay 5 times more for bandwidth to accommodate the additional frequency-range. (all things equal eg. same codecs being used).

* How many conversations, or part there of, are held above 3.4kHz :D

* I think microphones have come a long way since the carbon days... and what about noise cancellation.

* Digital transmission should also lead to less noise being introduced. (compared to the old analog systems)
 
Last edited:
So what stops telcos offering HD voice services? We seemingly have the broadband available so why are they so hesitant in exploiting high fidelity telephony? It is all tied up in the interconnect facilities, an issue which requires a radical make over.
Click to expand...
What a stupid statement.

Most network traffic stays on-net, so how does interconnect even come into the picture? It's all about spectrum capacity and servicing as many people as possible within those constraints. Many networks support half-rate for exactly this reason.

This author previously showed his ignorance and lack of understanding of many things technical, but this one is a gem.

Idiot.
 
TwoCents1000 said:
What a stupid statement.

Most network traffic stays on-net, so how does interconnect even come into the picture? It's all about spectrum capacity and servicing as many people as possible within those constraints. Many networks support half-rate for exactly this reason.

This author previously showed his ignorance and lack of understanding of many things technical, but this one is a gem.

Idiot.
Click to expand...

I wouldn't go to such extremes to bash, but definitely to add that:

"With the common restriction of 300 Hertz to 3400 Hertz, telephone systems only carriy 20% of the frequencies that are present in human speech. " <- Human speech safely ends at about 10 to 12 kHz (baseband ofcourse), the amount of power our vocal cords can produce above those frequencies is minimal if not negligible.

"In contrast, some modern business audio systems such as telepresence and next generation telephones and handsets will reproduce frequencies up to 22 kHz." <- That's because we can hear up to about 20 kHz (when we're young) so one needs to be able to reproduce those frequencies for music.

"Perhaps more common in the next generation systems is a bandwidth of 80 – 70,000 Hertz; double the audio bandwidth that most of us experience today." <- No way a human can hear anything past 22 kHz (best case scenario). You're probably confusing yourself with sampling frequency (which you need to be at least twice the maximum baseband frequency to have an alias-free reproduction [Shannon-Nyquist theorem]). There's a reason audio formats have sample rates like 44.1 kHz (think 22.05 x 2). The higher your sample frequency, the larger the space between samples (in frequency), the less effect your non-ideal filters have when reconstructing the samples.

And finally, the reason why we're still conversing using 2G (GSM) vocoders can be mainly attributed to the fact that it's the only thing those GSM basestations do nowadays (and EDGE/GPRS for the last few non-3G areas). If you want high quality, code & sample the voice for higher baseband bandwidth and transmit using 3G but this would then make your GSM BSes expensive paperweights and only increase traffic on your UMTS sector antennae.

As always, it's a business problem/decision rather than a technological one.
 
Last edited:
UnUnOctium said:
I wouldn't go to such extremes to bash, but definitely to add that:

"With the common restriction of 300 Hertz to 3400 Hertz, telephone systems only carriy 20% of the frequencies that are present in human speech. " <- Human speech safely ends at about 10 to 12 kHz (baseband ofcourse), the amount of power our vocal cords can produce above those frequencies is minimal if not negligible.

"In contrast, some modern business audio systems such as telepresence and next generation telephones and handsets will reproduce frequencies up to 22 kHz." <- That's because we can hear up to about 20 kHz (when we're young) so one needs to be able to reproduce those frequencies for music.

"Perhaps more common in the next generation systems is a bandwidth of 80 – 70,000 Hertz; double the audio bandwidth that most of us experience today." <- No way a human can hear anything past 22 kHz (best case scenario). You're probably confusing yourself with sampling frequency (which you need to be at least twice the maximum baseband frequency to have an alias-free reproduction [Shannon-Nyquist theorem]). There's a reason audio formats have sample rates like 44.1 kHz (think 22.05 x 2). The higher your sample frequency, the larger the space between samples (in frequency), the less effect your filters have when reconstructing the samples.

And finally, the reason why we're still conversing using 2G (GSM) vocoders can be mainly attributed to the fact that it's the only thing those GSM basestations do nowadays (and EDGE/GPRS for the last few non-3G areas). If you want high quality, code & sample the voice for higher baseband bandwidth and transmit using 3G but this would then make your GSM BSes expensive paperweights and only increase traffic on your UMTS sector antennae.

As always, it's a business problem/decision rather than a technological one.
Click to expand...

Maybe I came across strongly, but when someone calls himself an engineer and he can't get the basics right....or not bother to think about it....and then make an idiotic statement in a published article, I get hot under the collar. People like this are an insult to engineers everywhere.

Like the 70KHz statement in the article. Another statement showing complete lack of knowledge or logic. As you pointed out, humans can't even hear to these frequencies, why the hell would you want to waste bandwidth on encoding information no-one is going to need or hear? Maybe the author is a member of some canine or bat species.

Given a choice, most people would rather have the ability to actually make a call (even at lower audio quality) rather than having stunning Hi-Fi sound (from a 1cm-diameter speaker, nogal) but with half the chance of actually being able to make the call in the first place. (Because all the capacity is being used by a few with "HD phones")
 
Last edited:
Viber is using a HD audio codec, and it is sounding brilliant!

One of the best VOIP apps I have ever used.
 
chopsuey said:
Yip

* If we are only using 20% of the freq range as the article states, would people be prepared to pay 5 times more for bandwidth to accommodate the additional frequency-range. (all things equal eg. same codecs being used).

* How many conversations, or part there of, are held above 3.4kHz :D

* I think microphones have come a long way since the carbon days... and what about noise cancellation.

* Digital transmission should also lead to less noise being introduced. (compared to the old analog systems)
Click to expand...

good points,
and the good old carbon mike.
 
TwoCents1000 said:
Maybe I came across strongly, but when someone calls himself an engineer and he can't get the basics right....or not bother to think about it....and then make an idiotic statement in a published article, I get hot under the collar. People like this are an insult to engineers everywhere.

Like the 70KHz statement in the article. Another statement showing complete lack of knowledge or logic. As you pointed out, humans can't even hear to these frequencies, why the hell would you want to waste bandwidth on encoding information no-one is going to need or hear? Maybe the author is a member of some canine or bat species.

Given a choice, most people would rather have the ability to actually make a call (even at lower audio quality) rather than having stunning Hi-Fi sound (from a 1cm-diameter speaker, nogal) but with half the chance of actually being able to make the call in the first place. (Because all the capacity is being used by a few with "HD phones")
Click to expand...

Maybe you need to watch your temper, it is unstable. Yes journalist do confuse facts now and then, and that is where you come in. But if you want to make an intelligent statement yet using that kind of temper, then what we the readers take from that statement is, "Jeez what's up with this geek now" instead of taking to account what you trying to teach us.
 
TwoCents1000 said:
Maybe I came across strongly, but when someone calls himself an engineer and he can't get the basics right....or not bother to think about it....and then make an idiotic statement in a published article, I get hot under the collar. People like this are an insult to engineers everywhere.
Click to expand...

Agreed. People do make mistakes but there's way too many in this article and they're wrong at a fundamental level.

Robertvv said:
Cell phones divides this by 4
Click to expand...

Que? 1) 'Divide' what by 4? 2) ""With the common restriction of 300 Hertz to 3400 Hertz, ." is a quote from the article.
 
I would just like cell phone calls to have the same quality as land lines. It's always a nightmare trying to explain something technical to someone on a cell phone.
 
I honestly haven't noticed a difference in the quality between land lines and cellphones. On both I can usually hear the other person perfectly fine. Cellphones suffer from various forms of interference and antenna problems so for me if they could fix that then things would be perfect.

I also cannot see the point of having HD quality telephone connections, perhaps someone can enlighten me as to what they are good for. Perhaps those telemarketing people, or customer service folks would sound more intelligent and actually be able to correctly diagnose and fix a simple problem once and for all. If HD quality does that then I am game otherwise skip this nonsense and bring on video calling so the person on the other side can see the look of frustration on my face when I have to put with their crap.
 
Since when do bandwidth and frequency range have anything to do with each other on an analogue service?

80-70,000 Hz? Most people cannot hear above 16,000 Hz. More important still, it's important to discern between how much real bandwidth is available for any given frequency range. Think in MP3 terms;

96kbps for 44kHz audio means you have 96kbps bandwidth per channel at 22kHz frequency range (0-22,000 range) in stereo or 48kbps per channel at 22kHz frequency range per channel if quadrophonic, or 32kbps per channel at 22kHz in six channels for 5.1 audio.

The higher the bandwidth provided the higher the potential granularity for the audio decoded which improves the relative clarity of the audio. The frequency range increasing does not automatically mean that the quality improves; it can in fact mean the opposite.

I could go on at length with this, but either I simply didn't skim the article properly, the poster made a simple mistake or the poster or who they were quoting does not know what they are talking about.
 
TwoCents1000 said:
Most network traffic stays on-net, so how does interconnect even come into the picture? It's all about spectrum capacity and servicing as many people as possible within those constraints. Many networks support half-rate for exactly this reason.
Click to expand...

If you are an MTN subscriber, and MTN has around 30-35% market share, than statistically you may make most mobile calls off net. If you phone fixed lines, these are all off-net and if you phone international, it is all off-net. The amount of off-net traffic will increase as more voip players enter the market (as they are doing in South Africa)

The point is that one can use wonderful technologies such as high-definition voice, video, instant messaging but as soon as you cross networks, you lose much of the functionality and much of the calling cost goes towards the interconnect. That is why we are promoting the concept of neutral peering hubs, to allow new technologies to cross networks and so gain popularity and become the norm.

It's all about an ecosystem. You can't have a one-fax-machine network, or a high definition TV without high definition content. In the same way, the whole industry needs to embrace high-definition voice so it becomes the standard and everybody benefits from economies of scale. This vision of 'telco empowerment' is covered in the white papers and article on www.xconnect.net
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X