View Full Version : Hi fidelity SSB bandwith
kn4ds
03-19-2007, 01:43 AM
An interesting discussion of what happens to audio, including voice, as you reduce the bandwidth:
http://www.rwonline.com/referen....2.shtml (http://www.rwonline.com/reference-room/special-report/07_rwrf_jan_19_part_2.shtml)
My only concern with eSSB is that its use is considerate of other users.
ab0wr
03-19-2007, 01:57 AM
Quote[/b] (n5rfx @ Mar. 18 2007,14:32)]Using the NPR program from K3VR shows that not only widening the bandwidth can bring about intelligibility, but also narrowing the bandwidth may not degrade intelligibility. #The comparison I am making in the sample provided is first the 14K bandwidth with white noise giving a SNR of 9 to 10 dB. #The second part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. #It seems that both parts of the sample have the same intelligibility, but the second half is less than 25% of the bandwidth of the first part. #Also the second part has 6 dB per octave preemphasis. #The overall S/N improvement was 9 dB. #
NPR Audio Sample (http://home.comcast.net/~mdmiller7/noise/npr_compare.mp3)
73,
Mark N5RFX
Mark,
I'm not sure everyone caught on to what you were doing.
By doing the pre-emphasis on the second part of sample, you aren't giving quite a "fair" comparison. Not that it isn't quite revealing.
If you listen carefully to the second part there are some words that aren't nearly as clear -- for instance "show". I knew what she was saying from the context but if I had heard her saying that by itself, I would have had a very hard time distinguishing whether she said "show" or "so".
There is no doubt in my mind that preemphasis on the upper part of the audio *is* something that needs to be experimented with. It is what Joe and others have pointed out for the Heil series of microphones. That's probably why the HC-5 element that I have works so well.
My guess is, however, that if you had three sections in that clip, with the last one being 500hz-4000hz with a 6db per octave preemphasis you would find an even bigger difference for it.
tim ab0wr
ab0wr
03-19-2007, 02:13 AM
dpoQuote[/b] ]I look at it this way, for every three stations running just 4 KHZ audio that's one average ham that can't find a frequency.
It's not illegal, but it is rather thoughtless especially given that it's just something a very few like but is certainly not necessary for communications.
You still haven't gone out and looked up the metrics for spectrum efficiency and interference mitigation, have you?
I *KNOW* you could have three conversations going on on 3920khz at 6pm running 200watts/5khz signals with one conversation being on the east coast, one on the gulf coast, and one in the central US.
How many conversations do you suppose you could have in that 5khz with stations running 1000watts/2.5khz?
My guess is *TWO*. One on the east coast at 3920khz and one in the Central US on 3922.5khz.
That is *WORSE* spectrum efficiency and interference mitigation for the NARROW band operation.
If you gain enough intelligibility from a wider, preemphasized signal then you may also gain the ability to run less average power. That is GOOD for spectrum efficiency and interference mitigation instead of BAD as you would paint it.
There needs to be more experimentation done on the ham bands to verify that higher articulation ratio's for wider, pre-emphasized speech allows lower average power levels to be used. If it turns out to be true then the major roadblock to implementing a communications paradigm that actually provides for better operation will be "converting" those of the old faith that "2100hz is good enough".
Without the experimentation we'll never even be able to find out if would truly be a boon to traffic handlers carrying messages so the next ARRL test of the NTS might turn out better.
*That* would be a real shame.
tim ab0wr
ab0wr
03-19-2007, 02:17 AM
Brian,
I think your dissertation was pretty well on the mark.
Do we have your permission to quote you?
tim ab0wr
Of course you can quote me Tim. Save it and post it anywhere you want to. It's just something I knocked out this morning after spending several days reading some of the literature. It's amazing how many different research articles on bandwidth and intelligibility there are from so many diverse disciplines. All of them (including military telecom companies) discuss the benefits of increased audio bandwidth.
Dave, wider bandwidths are a tradition in ham radio, and the "selfishness and limited space fallacy" is a bunch of hogwash. All modes are created equal and no mode should ever take precedence over any other mode. Imagine if the CW guys and the PSK31 guys started telling the NSSB guys they were all being selfish? Can you imagine the crying and gnashing of teeth? Puhlease.
The Bottom Line: 20 guys using 6 kHz is much more efficient than 4 guys using the same 6 kHz. It's an incredibly lame argument. Count heads not Hertz! Look at the bottom end of 80m right now. There are many free spaces on 80m and 40m since the bands were reorganized, and 20m has never been a problem except in the minds of a few. Is ham radio a dying hobby/service or is it too crowded? It can't be both.
Experiment; have fun; use all the modes that are legally available -- and if our spectrum ever does get too crowded, we'll invent ways to deal with it. However, I'm going to predict that by the time we need more space, if we ever need more space, we'll have invented new protocols to take care of the problem.
Brian, K3VR
[edited for typos]
kn4ds
03-19-2007, 04:21 AM
Quote[/b] (K3VR @ Mar. 18 2007,23:14)]Dave, wider bandwidths are a tradition in ham radio, and the "selfishness and limited space fallacy" is a bunch of hogwash. All modes are created equal and no mode should ever take precedence over any other mode.
Not my point.
I do *not* accuse anyone of QRMing anyone. I only say that wider bandwidths are fine, so long as nobody's interfered with.
I "get" the idea of more QSOs with less power with more intelligibility... so long as DX can continue.
Of course DX can continue Rufus. You got the DX windows for DX. You got your contests for DX. You got a bunch of guys every other weekend screaming their brains out so they can yell 5/9 and collect another point or two. DX will always be here. I just looked at 80m and found 50 kHz of free space before I even got to the DX window.
And speaking of 80 meters -- did you hear 43 tonight? Now, that's a selfish use of bandwidth. I never heard a bigger bunch of idiots in my life. If you have crap like that on 80m (and we do, regularly) then you shouldn't criticize a bunch of well behaved guys who want to experiment with audio.
.
kn4ds
03-19-2007, 04:36 AM
Quote[/b] (K3VR @ Mar. 18 2007,23:32)]And speaking of 80 meters -- did you hear 43 tonight? Now, that's a selfish use of bandwidth. I never heard a bigger bunch of idiots in my life. If you have crap like that on 80m (and we do, regularly) then you shouldn't criticize a bunch of well behaved guys who want to experiment with audio.
Yeah, we're listening to 43 here right now... as mentioned in the other thread, it's kind of like a wreck... kind of can't tear myself away.
I know there's bandwidth available. No arguments here from me.
And I don't mean to criticize anybody experimenting with anything. My comment was not meant as criticism at all.
I thought the Radio World article I mentioned earlier was very good on the subject of bandwidth... while glossing over a lot, it was a good starting place for those who wonder about intelligibility.
Don't worry about it Dave, I didn't take offense. I'm going to read that article and then hit the hay. Where's Charlie when we need him? According to Charlie, if there's an issue worth railing about in amateur radio, it's the ARRL's desire to seed the CW/RTTY segments with automatic robots sending email from the internet so they can stay afloat as a business by demonstrating that hams are conversant with an interoperable digital platform that will net them big Homeland Security grants so Dave Sumner won't go hungry. Ask Charlie about it -- and 73 to all de K3VR.
.
n5rfx
03-19-2007, 12:12 PM
Quote[/b] (ab0wr @ Mar. 17 2007,19:57)]My guess is, however, that if you had three sections in that clip, with the last one being 500hz-4000hz with a 6db per octave preemphasis you would find an even bigger difference for it.
Tim,
Here is the revised clip (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3) with 500hz-4000hz with a 6db per octave preemphasis at the end.
1st part 14K bandwidth with white noise giving a SNR of 9 to 10 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 18 dB SNR
3rd part 500hz-4000hz with a 6db per octave preemphasis 15dB SNR
73,
Mark N5RFX
WA0LYK
03-19-2007, 02:07 PM
Quote[/b] (n5rfx @ Mar. 19 2007,05:12)]Quote[/b] (ab0wr @ Mar. 17 2007,19:57)]My guess is, however, that if you had three sections in that clip, with the last one being 500hz-4000hz with a 6db per octave preemphasis you would find an even bigger difference for it.
Tim,
Here is the revised clip (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3) with 500hz-4000hz with a 6db per octave preemphasis at the end.
1st part 14K bandwidth with white noise giving a SNR of 9 to 10 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 18 dB SNR
3rd part 500hz-4000hz with a 6db per octave preemphasis 15dB SNR
73,
Mark N5RFX
Mark,
The last sample sounded "pinched" to me. You could certainly pick out the 'sh' and 's' (plural), such as "shows" very well.
Would it be hard to add a 4th sample to the end with a response just like the third only from 300 to 4000 Hz?
The extra low frequencys might add a little 'presence'.
Thanks
Jim
WA0LYK
n5rfx
03-19-2007, 02:14 PM
Jim,
I had to repost the sample because I inadvertently lowered the sample rate which squeezed the bandwidth of the first and last samples. I will add your request.
73,
Mark N5RFX
n5rfx
03-19-2007, 02:23 PM
Here is the revised clip with 300hz-4000hz with a 6db per octave preemphasis at the end.
1st part 14K bandwidth with white noise giving a SNR of 9 to 10 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 18 dB SNR
3rd part 500hz-4000Hz with a 6db per octave preemphasis 15dB SNR
4th part 300-4000Hz with a 6db per octave preemphasis 10 dB SNR
Revised NPR samples (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3)
73,
Mark N5RFX
Interesting stuff. There are a few unknowns with clips taken from the internet: What kind of pre-emphasis has been applied in the studio, and; what kind of compression algorithms have been applied in the analog to digital conversion, and; how much information is lost in the conversion to mp3 format?
Also: Working with a male voice will give us a better idea about what happens when we cut off the low end, since women usually don't produce much vocal information below 200 Hz.
.
n5rfx
03-19-2007, 03:16 PM
With all of changes I made, I was afraid that the 4 samples were not correct, so I performed the analysis again.
1st part 14K bandwidth with white noise giving a SNR of 9 to 11 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 17 dB SNR
3rd part 500hz-4000Hz with a 6db per octave preemphasis 14dB SNR
4th part 300-4000Hz with a 6db per octave preemphasis 16 dB SNR
The SNR on that last sample is what alerted me to the fact that I had lost control. These all pass the sanity test.
NPR Bandwidth, Preemphasis, and Noise Comparison (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3)
73,
Mark N5RFX
n5rfx
03-19-2007, 03:34 PM
Quote[/b] (K3VR @ Mar. 18 2007,09:11)]Interesting stuff. There are a few unknowns with clips taken from the internet: What kind of pre-emphasis has been applied in the studio, and; what kind of compression algorithms have been applied in the analog to digital conversion, and; how much information is lost in the conversion to mp3 format?
Also: Working with a male voice will give us a better idea about what happens when we cut off the low end, since women usually don't produce much vocal information below 200 Hz.
.
Since all samples were derived from the same clip, the pre-emphasis, compression and mp3 loss should not affect the comparison. I agree if the comparisons were made with differing initial pre-emphasis, compression, and mp3 loss, then we would have a problem.
The female/male voice was interesting to me. Our speaker in these last samples still has most of her voice energy at 500Hz or less. That is why the SNR goes up when the lower cutoff frequency goes down. At 200 Hz our speaker looses a little over 4dB of voice energy.
200Hz 4 dB
300 Hz 2.5 dB
400 Hz 2 dB
500 Hz 1.7 dB
600 Hz 1.4 dB
73,
Mark N5RFX
You're right, the pre-emphasis and etc. won't affect what we have to work with. Her clip was not overly processed or compressed prior to conversion as far as I can tell. It's very natural sounding. Here's a link to Ted Koppel's podcast, which has been eq'd, but which also doesn't sound overly compressed prior to conversion:
Podcast (http://www.npr.org/rss/podcast/podcast_detail.php?siteId=5495239)
Choose the button that says "Download MP3/4"
Forgot to mention -- the research says we need 12 dB s/n ratio for excellent intelligibility, so keep the peak noise level 12 dB below the audio if possible.
Quote[/b] (ab0wr @ Mar. 18 2007,19:41)]Quote[/b] (N3JI @ Mar. 17 2007,19:33)]///SNIP///I don't have my punchy clip with me (as I mentioned earlier), but I'll go ahead and lay my eSSB audio clip out here. It was recorded off the RF link, not at the audio or IF level. What you hear is what's going out. I guarantee you don't hear teeth or tongues in my clip. Flame away... http://www.qrz.com/iB_html/non-cgi/emoticons/biggrin.gif
http://mysite.verizon.net/res1t267/Files/N3JI.mp3
Joe,
It sounds good to me and I don't even have my hearing aids in tonight!!
You say your bass goes down to 100hz? I didn't notice a lot of bass in the clip. Do you equalize it down or is your voice like mine without a lot of bass to begin with?
tim ab0wr
Tim,
Yes I have energy down to a bit below 100 Hz with a flat studio condenser mic and no processing. Here's what my voice looks like "naked". http://www.qrz.com/iB_html/non-cgi/emoticons/biggrin.gif
http://mysite.verizon.net/res1t267/Images/NaturalVoice2.jpg
Joe,
In the .jpg you attached, at 100 Hz you're only about 7 dB down from the peak energy in that particular syllable. I hear lots of low end in that clip.
I found my "Punchy" clip I've been talking about! It's embedded in a PowerPoint slide, so you'll need that to hear it. Double-click the speaker icon to listen. The slide includes a spectrum plot of the clip as well:
http://mysite.verizon.net/res1t267/Files/N3JI_Punchy.ppt
Joe, N3JI
Quote[/b] (K3VR @ Mar. 19 2007,09:57)]Joe,
In the .jpg you attached, at 100 Hz you're only about 7 dB down from the peak energy in that particular syllable. I hear lots of low end in that clip.
Yeah, I know. I have EQ'ed it for a good amount of low end, so it's there. As opposed to the other clip I just posted...
http://www.qrz.com/iB_html/non-cgi/emoticons/smile.gif
Quote[/b] (n5rfx @ Mar. 19 2007,08:23)]Here is the revised clip with 300hz-4000hz with a 6db per octave preemphasis at the end.
1st part 14K bandwidth with white noise giving a SNR of 9 to 10 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 18 dB SNR
3rd part 500hz-4000Hz with a 6db per octave preemphasis 15dB SNR
4th part 300-4000Hz with a 6db per octave preemphasis 10 dB SNR
Revised NPR samples (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3)
73,
Mark N5RFX
I'm repeating myself, but this is excellent work. It simulates twisting the BW and/or IF shift knobs on your rig, and I still prefer the 300-4k clip, with the 200-3k & 500-4k being about tied for next. Although the S/N ratio is better using 2.8k, the loss of the top end is quite noticeable.
I still prefer the ~4k range as a good compromise between BW & intelligibility. That's probably why I'm happy with my 4.2k transmit BW.
Thanks again for some nice editing work.
Joe, N3JI
n5rfx
03-19-2007, 04:52 PM
Quote[/b] (N3JI @ Mar. 18 2007,10:40)]I still prefer the ~4k range as a good compromise between BW & intelligibility.
Joe,
I am glad to be able to analyze these samples. I agree that 4.5 KHz seems to be a good compromise. From 2.8k to 4.5 k you only add 2 dB to the noise, so that should not cause any problems.
I will work on Ted Koppel's podcast later today using the same setup. I am running out of space on the server I have been using, so some of the links to other samples will start disappearing as we add more.
73,
Mark N5RFX
Yep, that's punchy alright Joe. Contest audio for sure! Thanks for sharing it. And thanks for playing with the other clips too Mark. All interesting stuff for sure.
Quote[/b] (K3VR @ Mar. 19 2007,11:19)]Yep, that's punchy alright Joe. Contest audio for sure! Thanks for sharing it. And thanks for playing with the other clips too Mark. All interesting stuff for sure.
BTW, I'm chopping at ~3400 for two reasons: One is I don't think too many folks listen wide enough even with the IF shifted to hear beyond that, and the second is simply to keep the QRM down. #So I could go almost another full kHz, but I'm sure my neighbors on the crowded band wouldn't care for it.
Joe, N3JI
ab0wr
03-19-2007, 07:36 PM
Quote[/b] (n5rfx @ Mar. 19 2007,08:34)]Quote[/b] (K3VR @ Mar. 18 2007,09:11)]Interesting stuff. There are a few unknowns with clips taken from the internet: What kind of pre-emphasis has been applied in the studio, and; what kind of compression algorithms have been applied in the analog to digital conversion, and; how much information is lost in the conversion to mp3 format?
Also: Working with a male voice will give us a better idea about what happens when we cut off the low end, since women usually don't produce much vocal information below 200 Hz.
.
Since all samples were derived from the same clip, the pre-emphasis, compression and mp3 loss should not affect the comparison. #I agree if the comparisons were made with differing initial pre-emphasis, compression, and mp3 loss, then we would have a problem.
The female/male voice was interesting to me. #Our speaker in these last samples still has most of her voice energy at 500Hz or less. #That is why the SNR goes up when the lower cutoff frequency goes down. #At 200 Hz our speaker looses a little over 4dB of voice energy.
200Hz 4 dB
300 Hz 2.5 dB
400 Hz 2 dB
500 Hz 1.7 dB
600 Hz 1.4 dB
73,
Mark N5RFX
Very nice.
Did anyone else notice that they could not tell whether the speaker is saying "pattern" or "patterns" in either of the first two clips?
The first clip apparently has too low of an SNR, probably because of the wide bandwidth. The second clip just doesn't have enough high freq content to clearly make out the "s" part of the word.
Yet with the 4khz clips I had no problem telling she was saying "patterns".
While I like the sound of the 300hz to 4000hz clip better I haven't been able to tell much improvement over the 500hz to 4000hzz clip. You certainly don't get a good "feel" for how she would sound in real life from the 500hz clip.
Does anyone else notice that or is it just my ears. I do have my hearing aids in. I didn't try listening without them, perhaps I'll do that and see how it sounds.
tim ab0wr
ab0wr
03-19-2007, 07:42 PM
Quote[/b] (ab0wr @ Mar. 19 2007,12:36)]Quote[/b] (n5rfx @ Mar. 19 2007,08:34)]Quote[/b] (K3VR @ Mar. 18 2007,09:11)]Interesting stuff. There are a few unknowns with clips taken from the internet: What kind of pre-emphasis has been applied in the studio, and; what kind of compression algorithms have been applied in the analog to digital conversion, and; how much information is lost in the conversion to mp3 format?
Also: Working with a male voice will give us a better idea about what happens when we cut off the low end, since women usually don't produce much vocal information below 200 Hz.
.
Since all samples were derived from the same clip, the pre-emphasis, compression and mp3 loss should not affect the comparison. #I agree if the comparisons were made with differing initial pre-emphasis, compression, and mp3 loss, then we would have a problem.
The female/male voice was interesting to me. #Our speaker in these last samples still has most of her voice energy at 500Hz or less. #That is why the SNR goes up when the lower cutoff frequency goes down. #At 200 Hz our speaker looses a little over 4dB of voice energy.
200Hz 4 dB
300 Hz 2.5 dB
400 Hz 2 dB
500 Hz 1.7 dB
600 Hz 1.4 dB
73,
Mark N5RFX
Very nice.
Did anyone else notice that they could not tell whether the speaker is saying "pattern" or "patterns" in either of the first two clips?
The first clip apparently has too low of an SNR, probably because of the wide bandwidth. The second clip just doesn't have enough high freq content to clearly make out the "s" part of the word.
Yet with the 4khz clips I had no problem telling she was saying "patterns".
While I like the sound of the 300hz to 4000hz clip better I haven't been able to tell much improvement over the 500hz to 4000hzz clip. You certainly don't get a good "feel" for how she would sound in real life from the 500hz clip.
Does anyone else notice that or is it just my ears. I do have my hearing aids in. I didn't try listening without them, perhaps I'll do that and see how it sounds.
tim ab0wr
Ok, without my hearing aids I had a problem hearing the "s" in patterns in all of the clips. My ears, by themselves, make pretty good low pass filters!!!
I put them back in and I can hear the "s" in the last two clips but not in the first two, at least not very well.
If someone was using that word in a traffic message it would be very easy for an intermediate relay person to write down "pattern" instead of "patterns". I know it doesn't sound like much but I keep going back to the old "want of a nail" poem!
tim ab0wr
n5rfx
03-19-2007, 08:19 PM
Now for Ted Koppel. His mail voice does make a difference in the occupied spectrum.
LPF Cut Off Freq.(Hz) Female (dB) Male (dB)
200 4.24 6.02
300 2.53 3.56
400 2.00 1.68
500 1.65 1.05
600 1.43 0.85
Both male and female speakers voice frequencies are primarily above 200 Hz. The female's frequencies seem to be more evenly distributed than the mail voice. The male voice has a significant amount of spectral density starting at about 300 Hz. In the Ted Koppel sample, the frequency range is limited to 7kHz. This is how I got the file from NPR. I added white noise (0 to 7K) to give a 12 dB SNR.
There are 4 parts to the Ted Koppel sample.
1st part: A 60 second sample of the original file with 12 dB SNR.
2nd part: A 60 second sample of the original file with 6 dB per octave pre-emphasis. The SNR is 14 dB
3rd part: A 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 4500 Hz filtering. The SNR is 16 dB
4th part: A 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 3000 Hz filtering. The SNR is 17 dB
Ted Koppel Sample (http://home.comcast.net/~mdmiller7/noise/npr_ted_compare.mp3)
We have not been considering the affects of AGC. With strong signals, the AGC may be able to drive the noise down below the thermal noise of the receiver. Conversely it is possible that the AGC will not be able to drive the noise below thermal as the bandwidth is widened because of an increase in bandwidth. With only a 2 dB increase in noise (calculated) from 2.8 to 4.5 KHz bandwidth I don't see this as being a problem, and as long as there is sufficient AI gain to the listener, then it is a worthwhile exercise. It is the AI gain that is the most subjective part of the analysis. I really don't have a feel for AI and what differences are significant. That will require some more reading.
73,
Mark N5RFX
n5rfx
03-19-2007, 08:31 PM
Quote[/b] (ab0wr @ Mar. 18 2007,13:36)]Did anyone else notice that they could not tell whether the speaker is saying "pattern" or "patterns" in either of the first two clips?
I heard patterns in all of the clips, but that may be because I heard it first from you. I do recall hearing patterns before you brought it up, because I remember thinking I wanted to look that up in the Internet. I am a Rod Serling fan. My eyes started to go about 3 years ago, and I am waiting for the hearing to go next. I definitely have a family history of hearing loss. Oh boy!
73,
Mark N5RFX
wd0ct
03-19-2007, 08:47 PM
Part 1 dull and lifeless. Just like my tv audio on speech.
The tv has bass and treble boost available from 0 to 100, whatever that means.
It also has audio presets called
speech - 20 and 20
theater - 90 and 90
concert - 50 and 50
personal - I set around 60 and 70
Part 2 highly intelligible but too much for me.
Part 3 my favorite
Part 4 hammie audio
wd0ct
03-19-2007, 08:57 PM
After I posted the tv presets I thought about them again.
The theater setting of 90 and 90 does give a reasonable impression [considering the two little built in speakers] of the big audio I hear at a movie theater.
I don't get HBO but I can sometimes see the pic and hear the audio somewhat. The signal to noise is way worse than 12db though.
It is perfect for playing with the bass and treble controls for max intelligibility. I end up around 60 and 70 every time. Strictly high boost does not work for me.
EDIT: This was in reference to Tim's posts about the NPR clip, not the latest "Koppel" posts. I didn't refresh before replying...
Definite difference in the 500 HP & 300 HP clips, and most of the high freq stuff was chopped with the 3k LP, so nothing earth shattering there. #I did find the full-range clip annoying, though I didn't pay particular attention to any lost sibilants due to noise. #That's why I chop as low as I can without killing the highs.
And here I thought you were kidding about the hearing aide comment a few posts ago! #Try listening to my eSSB clip with your HAs in and see if you notice any difference in the lows. #There are plenty there, so I'm wondering if your ears or perhaps the HAs are chopping the lows.
Joe, N3JI
WA0LYK
03-19-2007, 09:39 PM
Quote[/b] (N3JI @ Mar. 19 2007,09:40)]Quote[/b] (n5rfx @ Mar. 19 2007,08:23)]Here is the revised clip with 300hz-4000hz with a 6db per octave preemphasis at the end.
1st part 14K bandwidth with white noise giving a SNR of 9 to 10 dB.
2nd part of the sample reduces the bandwidth from 14K to 2.8K (200 to 3000) Hz. 18 dB SNR
3rd part 500hz-4000Hz with a 6db per octave preemphasis 15dB SNR
4th part 300-4000Hz with a 6db per octave preemphasis 10 dB SNR
Revised NPR samples (http://home.comcast.net/~mdmiller7/noise/npr_compare_1.mp3)
73,
Mark N5RFX
I'm repeating myself, but this is excellent work. It simulates twisting the BW and/or IF shift knobs on your rig, and I still prefer the 300-4k clip, with the 200-3k & 500-4k being about tied for next. Although the S/N ratio is better using 2.8k, the loss of the top end is quite noticeable.
I still prefer the ~4k range as a good compromise between BW & intelligibility. That's probably why I'm happy with my 4.2k transmit BW.
Thanks again for some nice editing work.
Joe, N3JI
I agree, the last clip (300-4000) sounds the best. The 500-4000 Hz is just missing too much. The 200-2800 would be my second choice.
Looking at the frequency response of the clip with xmms, it looks like the preemphasis makes the response flat, i.e. the highs show as much loudness as the lows. It sure lets you pick out some of the sounds that the first clip smears. My hearing isn't the best, and maybe this is just making up for that. It may even be making up for poor response of my little Sony speakers. In any case, the intelligibility is better when the highs are boosted.
Now, about the Koppel clip...
I found cutting at 200 removed a significant amount of his "personality". But clearly, the 200-4.5k is the most intelligible.
Joe
The upper register pre-emphasis seems to help intelligibility by raising the energy level of certain consonants. It's not pleasant listening but it helps. The extended bandwidth in the upper register always helps intelligibility. I don't lose any intelligibility with the lows chopped but I do lose the intention and emotional tone of the speaker.
It's interesting how much subtlety of intonation and personality is in the lower portion of the male register. I've been listening to the way actors portray emotion in books on tape. Anger, for instance, usually comes across as a low growl; fear is portrayed in the higher registers; shock is often whispered. These guys make a living imitating human behavior. It makes me wonder if intonation in language is unconciously paired with a particular frequency for a given emotion or intention.
.
ab0wr
03-20-2007, 01:20 AM
Quote[/b] (n5rfx @ Mar. 19 2007,13:19)]Now for Ted Koppel. #His mail voice does make a difference in the occupied spectrum.
LPF Cut Off Freq.(Hz) Female (dB) Male (dB)
200 # # # # # # # # # # # # 4.24 # # # # # #6.02 #
300 # # # # # # # # # # # # 2.53 # # # # # #3.56
400 # # # # # # # # # # # # 2.00 # # # # # #1.68
500 # # # # # # # # # # # # 1.65 # # # # # #1.05
600 # # # # # # # # # # # # 1.43 # # # # # #0.85
Both male and female speakers voice frequencies are primarily above 200 Hz. #The female's frequencies seem to be more evenly distributed than the mail voice. #The male voice has a significant amount of spectral density starting at about 300 Hz. #In the Ted Koppel sample, the frequency range is limited to 7kHz. #This is how I got the file from NPR. #I added white noise (0 to 7K) to give a 12 dB SNR.
There are 4 parts to the Ted Koppel sample.
1st part: #A 60 second sample of the original file with 12 dB SNR.
2nd part: A 60 second sample of the original file with 6 dB per octave pre-emphasis. #The SNR is 14 dB
3rd part: A 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 4500 Hz filtering. #The SNR is 16 dB
4th part: A 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 3000 Hz filtering. #The SNR is 17 dB
Ted Koppel Sample (http://home.comcast.net/~mdmiller7/noise/npr_ted_compare.mp3)
We have not been considering the affects of AGC. #With strong signals, the AGC may be able to drive the noise down below the thermal noise of the receiver. #Conversely it is possible that the AGC will not be able to drive the noise below thermal as the bandwidth is widened because of an increase in bandwidth. #With only a 2 dB increase in noise (calculated) from 2.8 to 4.5 KHz bandwidth I don't see this as being a problem, and as long as there is sufficient AI gain to the listener, then it is a worthwhile exercise. #It is the AI gain that is the most subjective part of the analysis. #I really don't have a feel for AI and what differences are significant. #That will require some more reading. #
73,
Mark N5RFX
Mark,
One quick note, in the future it might work better to put the narrow bandwidth clips first. That makes it easier to spot areas of increased intelligibility since you haven't already been "biased" by the good clips.
Now, I immediately noticed that the word "inexperienced" was missing the "d" sound on the end in both the first and the last clip, at least for me. I need to wait a while before listening again so I can get the repeating loop out of my head so I can listen for other "missing" parts.
It would be extremely interesting to see the difference between clip two and three if the SNR for clip three went clear down to 14db. I'm pretty sure if you took clip 4 down to 14db there would be several areas that I would have a hard time with. My guess is that clip 3 at 14db will have lower intelligibility than clip 2, how much less I don't know.
tim ab0wr
n5rfx
03-20-2007, 07:16 AM
Quote[/b] (ab0wr @ Mar. 18 2007,19:20)]It would be extremely interesting to see the difference between clip two and three if the SNR for clip three went clear down to 14db. I'm pretty sure if you took clip 4 down to 14db there would be several areas that I would have a hard time with. My guess is that clip 3 at 14db will have lower intelligibility than clip 2, how much less I don't know.
tim ab0wr
Here it is:
The first part is a 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 4500 Hz filtering. The SNR is 14 dB
The second part is A 60 second sample of the original file with 6 dB per octave pre-emphasis. The SNR is 14 dB
Tim's Request (http://home.roadrunner.com/~mdmiller7/npr_ted_compare_tim.mp3)
73,
Mark N5RFX
n5rfx
03-20-2007, 07:31 AM
When pre-emphasis is added to the original 60 second sample, the PEP to average ratio goes up by 6 dB.
No pre-emphasis 15 dB
Pre-emphasis 21 dB
This ratio holds throughout the bandpass filtering.
73,
Mark N5RFX
WA0LYK
03-20-2007, 12:37 PM
Quote[/b] (n5rfx @ Mar. 20 2007,00:16)]Quote[/b] (ab0wr @ Mar. 18 2007,19:20)]It would be extremely interesting to see the difference between clip two and three if the SNR for clip three went clear down to 14db. I'm pretty sure if you took clip 4 down to 14db there would be several areas that I would have a hard time with. My guess is that clip 3 at 14db will have lower intelligibility than clip 2, how much less I don't know.
tim ab0wr
Here it is:
The first part is a 60 second sample of the original file with 6 dB per octave pre-emphasis and 200 - 4500 Hz filtering. The SNR is 14 dB
The second part is A 60 second sample of the original file with 6 dB per octave pre-emphasis. The SNR is 14 dB
Tim's Request (http://home.roadrunner.com/~mdmiller7/npr_ted_compare_tim.mp3)
73,
Mark N5RFX
Some comments.
I had no problem understanding everything said even at this noise level. The noise is starting to become distracting but I think preemphasis would help at an even higher level of noise.
The preemphasis on the unfiltered clip made the s's sound overemphasized, almost like a snake. Might have to live with this to get the other benefits. There would be no doubts as to what words were plural ('s' on the end). I didn't notice this at all on the filtered clip.
The 'b' sound in the word 'bloated' was much more pronounced in the unfiltered clip. I don't know if going down to 100 Hz would help a lot with the AI since it was still pretty easy to understand what was being said. However, 'easy' listening would be enhanced with a few more low freqs.
Some thoughts.
It would be nice to hear what some folks think that have both good hearing and a good receive audio setup on their rigs. Is the preemphasis 'over the top' for these folks? I know it must be making up some for my hearing loss and for not poor but not the best speakers on my pc. In other words, does it give the same benefit in articulation for these folks as for me? Is it destructive or beneficial for everyone?
It's funny how were talking about a bandwidth of what traditionally has been a good signal on AM. Did the first hams arrive at this bandwidth because it gave the most understandable signal through experimentation or because the equipment they had, i.e. microphones and audio amps just naturally fell into this range?
Jim
WA0LYK
n5rfx
03-20-2007, 01:36 PM
Quote[/b] (WA0LYK @ Mar. 19 2007,06:37)]The preemphasis on the unfiltered clip made the s's sound overemphasized, almost like a snake. Might have to live with this to get the other benefits. There would be no doubts as to what words were plural ('s' on the end). I didn't notice this at all on the filtered clip.
Jim,
One thing we have not been considering is de-emphasis on the receive end. Here is a clip with 3 samples.
1st part: A 60 second sample of the original file with 200 to 4500Hz bandpass filtering. The SNR is 12 dB
2nd part: A 60 second sample of the original file with 6dB per octave pre-emphasis and 200 to 4500Hz bandpass filtering. The SNR is 14 dB
3rd part: The second part with 6 dB per octave de-emphasis. The SNR is 19 dB
pre-emphasis_de-emphasis (http://home.roadrunner.com/~mdmiller7/compare_emphasis.mp3)
73,
Mark N5RFX
ab0wr
03-20-2007, 01:37 PM
Quote[/b] (N3JI @ Mar. 19 2007,14:15)]EDIT: This was in reference to Tim's posts about the NPR clip, not the latest "Koppel" posts. #I didn't refresh before replying...
Definite difference in the 500 HP & 300 HP clips, and most of the high freq stuff was chopped with the 3k LP, so nothing earth shattering there. #I did find the full-range clip annoying, though I didn't pay particular attention to any lost sibilants due to noise. #That's why I chop as low as I can without killing the highs.
And here I thought you were kidding about the hearing aide comment a few posts ago! #Try listening to my eSSB clip with your HAs in and see if you notice any difference in the lows. #There are plenty there, so I'm wondering if your ears or perhaps the HAs are chopping the lows.
Joe, N3JI
I have listened to the clip with the HA's in. Neither they or my ears have any problems with anything in the bass range.
It's just that you don't have a bass voice like Tennessee Ernie Ford. Most people don't, even announcers. You *do* hear that on the ham bands sometimes. I suspect because as someone pointed out (perhaps you) that some mic's will cause that because of their response characteristics.
tim ab0wr
Quote[/b] (WA0LYK @ Mar. 20 2007,06:37)]Some comments.
I had no problem understanding everything said even at this noise level. #The noise is starting to become distracting but I think preemphasis would help at an even higher level of noise. #
The preemphasis on the unfiltered clip made the s's sound overemphasized, almost like a snake. #Might have to live with this to get the other benefits. #There would be no doubts as to what words were plural ('s' on the end). #I didn't notice this at all on the filtered clip.
The 'b' sound in the word 'bloated' was much more pronounced in the unfiltered clip. #I don't know if going down to 100 Hz would help a lot with the AI since it was still pretty easy to understand what was being said. #However, 'easy' listening would be enhanced with a few more low freqs.I agree with pretty much everything you said here. #I find myself hardly ever cutting the bottom end, other than when the noise & QRM are at extreme levels. #For my operating habits, that seems to be during contests/pileups when I'm *really* trying for some rare station that's down in the dirt. #The preemphasis is definitely very bright, but when the noise levels increase, it sure punches though! #In fact, I'll probably have another look at my "punchy" settings to see if cranking the highs up even more will be of benefit.
Quote[/b] ]Some thoughts.
It would be nice to hear what some folks think that have both good hearing and a good receive audio setup on their rigs. #Is the preemphasis 'over the top' for these folks? #I know it must be making up some for my hearing loss and for not poor but not the best speakers on my pc. #In other words, does it give the same benefit in articulation for these folks as for me? #Is it destructive or beneficial for everyone?I have a pretty decent receive audio setup in my shack, but most of the time the guys I ragchew with have fairly balanced audio; not too many lows, and just the right amount of highs. #Only occasionally have I heard someone with what I consider "over the top" highs, and that's usually when they're just getting started, changed a lot of things around, got a new mic, etc. #That said, I'm listening to these clips on two different PCs. #I'm using Boston Acoustic speakers with a subwoofer on one and very good headphones on the other. #Similar results with both.
Quote[/b] ]It's funny how were talking about a bandwidth of what traditionally has been a good signal on AM. #Did the first hams arrive at this bandwidth because it gave the most understandable signal through experimentation or because the equipment they had, i.e. microphones and audio amps just naturally fell into this range?
Jim
WA0LYKI'm not sure what the answer to this is. #Since carbon mics were the first ones used I believe, I wonder what their natural frequency response is. #I've never really thought about the initial BW used on AM, but I'm betting it was limited by those old mics. #I could be wrong -- I'm just going on gut feel here...
Joe, N3JI
Quote[/b] (ab0wr @ Mar. 20 2007,07:37)]I have listened to the clip with the HA's in. Neither they or my ears have any problems with anything in the bass range.
It's just that you don't have a bass voice like Tennessee Ernie Ford. Most people don't, even announcers. You *do* hear that on the ham bands sometimes. I suspect because as someone pointed out (perhaps you) that some mic's will cause that because of their response characteristics.
tim ab0wr
LOL... #Okay, okay... #I know I don't have THIS VOICE (http://www.nu9n.com/mp3/N2VU_Donald.mp3).
But I'm not squeaky, either.
http://www.qrz.com/iB_html/non-cgi/emoticons/laugh.gif
Quote[/b] (K1VSK @ Feb. 27 2007,11:51)]I admit I just don't get it.
I just saw a pic of someone's shack posted here on QRZ.com showing a small rig and lots of audio stuff, all of which results is an audio signal wider than a barn door and of much less quality than traditional AM.
Given that AM is still widely used, what possible reason(s) are there for producing a poor substitute via processed SSB which only serves to use excessive spectrum?
As a paranthetical question, this same picture contained a bio written in the third person - are we to infer someone else wrote his boi?
Your correct, you DONT get it. #We enjoy sounding how we want to sound. IF your rcvr doesn’t open up to 6khz, and I’m pretty sure it doesn’t, then you cannot hear the fidelity. #My rack may cost more than most HF rigs but to me it's worth it.
K7OP
ab0wr
03-20-2007, 05:48 PM
Quote[/b] (n5rfx @ Mar. 20 2007,06:36)]Quote[/b] (WA0LYK @ Mar. 19 2007,06:37)]The preemphasis on the unfiltered clip made the s's sound overemphasized, almost like a snake. #Might have to live with this to get the other benefits. #There would be no doubts as to what words were plural ('s' on the end). #I didn't notice this at all on the filtered clip.
Jim,
One thing we have not been considering is de-emphasis on the receive end. #Here is a clip with 3 samples.
1st part: A 60 second sample of the original file with 200 to 4500Hz bandpass filtering. #The SNR is 12 dB
2nd part: A 60 second sample of the original file with 6dB per octave pre-emphasis and 200 to 4500Hz bandpass filtering. #The SNR is 14 dB
3rd part: The second part with 6 dB per octave de-emphasis. #The SNR is 19 dB
pre-emphasis_de-emphasis (http://home.roadrunner.com/~mdmiller7/compare_emphasis.mp3)
73,
Mark N5RFX
Boy, this is a tough one!
I *like* the sound of the middle clip better even though the noise is higher than on the last clip.
I'm not sure I can tell any difference in the intelligibility between the middle clip and the last clip, however.
I *still* can't hear the "d" on inexperienced at this bandwidth while I can on the 7khz clips.
I suspect the only way to really separate the intelligibility issues is to actually record some test syllables, words, and sentences and then actually play them for test subjects and record the results.
tim ab0wr
A little bit about the history of preemphasis in AM radio:
In the 80's, AM broadcast receivers were being designed with narrower and narrower filters and purposely limited audio response to prevent reception of interference at night as propagation changed. AM broadcasters were upset about this manufacturing trend, because they said it meant people were migrating to the FM band in search of the greater fidelity they were used to.
In 1986, responding to their concerns, the National Radio Systems Committee (NRSC) proposed a standard AM transmission response with a sharp audio cutoff at just under 10 kHz along with a standard pre-emphasis curve for stations operating in the AM broadcast band. #The standard was designed to allow the manufacturers of radios to increase the bandwidth of receivers for improved audio response while allowing radio stations to standardize bandwidth for acceptable fidelity while lowering the received interference from first adjacent stations in low signal areas.
"Many receivers using ceramic filters with narrow response characteristics can not be improved by use of excessive preemphasis. These receivers can not hear the transmission of preemphasized high audio frequencies. But excessive preemphasis will foster adjacent channel interference and cause wideband radios to sound shrill or strident." [NRSC-1]
"Audio response is best measured by detecting the over-the-air signal, This will ensure that the AM transmitter and antenna combination is faithfully reproducing the preemphasized audio." [NRSC-1]
"Some AM stations with transmitter or antenna problems may not be able to pass preemphasized audio without introducing "splatter" interference and/or overmodulation. If a station transmission system cannot "handle" the NRSC recommended curve, it is suggested that a lower amount of preemphasis be used until the problems are corrected to allow the NRSC curve to be faithfully implemented." [NRSC-1]
A corresponding deemphasis curve was proposed for the manufacture of AM radio receivers that would match the preemphasis of the AM stations. "...the implementation of preemphasis/deemphasis standards produces transmission/reception system that is essentially flat to nearly 10 kHz #and limited only by the AM #receiver's choice of bandwidth." [NRSC-1]
"Selectivity and Frequency Response [of Receivers]: Narrowband #50 Hz to 5000 Hz +/- #X dB; Wideband #50 Hz to 10,000 Hz +/- #X dB (Where X is the maximum deviation from the recommended frequency response, and 5000 Hz and 10,000 Hz are example frequency specifications.) The deviation X shall be of as low a value as practical." [NRSC-1]
NRSC-1 Was adopted by the FCC in 1990. See the NRSC document below for the preemphasis/deemphasis figures in use on the AM Broadcast band.
NRSC STANDARD (http://www.nrscstandards.org/Standards/nrsc-1.pdf)
Hints, Tips, and Recommendations for Amateurs Experimenting With Increased Fidelity/Intelligibility SSB
1. A receiver section employing a crystal filter system like a Collins KWM-2 will not hear any bandwidth change beyond 2,700 Hz, or below 300 Hz, by virtue of the "brick wall" filtering system. Many radios in use on the amateur bands today will be "deaf" to improvements in audio fidelity and intelligibility due to receiver design parameters.
2. In contrast to #1, a receiver like that found in an Icom 756 Pro III can detect and appreciate bandwidth changes between 100 Hz and 3,600 Hz. Therefore, improvements in intelligibility and fidelity can be detected out to the maximum receive bandwidth of 3,600 Hz in SSB. Likewise, a Kenwood TS-850 with DSP-100 can detect bandwidth changes resulting in improved intelligibility out to approximately 6 kHz. Many other radios currently in production can also detect increased bandwidths.
3. An improperly adjusted carrier point in a radio like a Yaesu MkV will cause an increase of transmitted energy on the unwanted sideband, adding nothing to fidelity or intelligibility, but vastly increasing interference potential to other spectrum users.
4. Too much audio preemphasis may cause IM3 and IM5 products to increase, causing potential for interference. It is the transmitting station's responsibility to monitor IMD carefully.
5. Receiving stations with noise blankers engaged will often complain of "splatter," when in fact, the noise blanker is adding unwanted artifacts to the recovered audio in the receiver.
6. Stations operating receivers in close proximity to other stations will often complain of "splatter" when in fact they are simply attempting to operate within the transmitted bandwidth envelope of the other station's transmitter. Let them know your transmitted bandwidth is wider than 3 kHz. Determine whether or not a frequency change is in your best interest.
7. The goal of achieving fidelity is sometimes at odds with the goal of achieving intelligibility. Because the energy present in the human voice drops off as frequency increases, preemphasis may be required to bring the level of certain consonants to a point where other stations (with appropriate equipment) can hear them with the desired intelligibility. Preemphasis decreases fidelity but it may improve intelligibility, particularly in high noise environments.
8. High Fidelity audio has traditionally represented a flat audio response, between zero and 20,000 Hz. Audio transmitted in narrower bandwidths may benefit from the use of preemphasis with the goal of improving intelligibility when wider bandwidths are impractical. The benefits of higher fidelity transmissions are only practical when, A.) Space is available for such transmissions without causing interference, and, B.) When atmospheric noise levels are low enough for the receiving station(s) to appreciate increases in fidelity.
9. Amateur stations employing preemphasis should use sufficient detection methods (oscilloscopes, spectrum analyzers) to be certain their transmitters are not the cause of harmful interference and spurious emissions. [See note] Any kind of audio processing, compression, or equalization scheme, including the predetermined +4dB rise in many microphone elements could conceivably contribute to harmful interference. Ensuring the transmission of a 'clean signal' should be considered a primary part of "good amateur practice."
10. Any introduction of nonlinearity (overload to the point of distortion) to the transmission system, whether in audio or RF amplification, will potentially result in greater interference potential. Grid driven tubes or improperly filtered 12v transistorized amplifiers, for example, are inherently less linear than amplifiers driven by triodes. The RF output of amplifiers should also be monitored.
11. Improper tuning of linear amplifiers, overuse of preemphasis, introduction of nonlinear audio, overuse of compression and/or processing, "turning up" the power output of amateur transceivers (creating nonlinearity) in order to provide higher drive for RF amplifiers, and RF feedback (antenna problems) will also present greater interference potential.
12. Increased low frequency response means that stations who are even 10 Hz "off frequency" from the receiving station will sound unnatural. Stations experimenting with transmitting low frequencies (under 100 Hz) are encouraged to employ transmitters using strict frequency tolerances. A TCXO referenced to WWV is highly recommended.
The Bottom Line: Major increases in SSB intelligibility and fidelity are possible. The responsibility for interference mitigation and running a clean station rests solidly with the station doing the transmitting. Amateur stations experimenting with higher fidelity/intelligibility must ensure that their transmissions comply with accepted standards of spectral purity.
[Note] Spurious emissions include harmonic emissions, parasitic emissions, intermodulation products and frequency conversion products. Make sure your station's transmitted signal complies with FCC rules!
.
n5rfx
03-21-2007, 06:43 PM
Quote[/b] (n5rfx @ Mar. 19 2007,07:36)]One thing we have not been considering is de-emphasis on the receive end. Here is a clip with 3 samples.
1st part: A 60 second sample of the original file with 200 to 4500Hz bandpass filtering. The SNR is 12 dB
2nd part: A 60 second sample of the original file with 6dB per octave pre-emphasis and 200 to 4500Hz bandpass filtering. The SNR is 14 dB
3rd part: The second part with 6 dB per octave de-emphasis. The SNR is 19 dB
pre-emphasis_de-emphasis (http://home.roadrunner.com/~mdmiller7/compare_emphasis.mp3)
Here are some 1/3 octave band displays (http://home.roadrunner.com/~mdmiller7/compare_emphasis.jpg) for the 3 samples in the pre-emphasis_de-emphasis (http://home.roadrunner.com/~mdmiller7/compare_emphasis.mp3) clip. The 4th display is a reference display that shows the sample as it was downloaded off of the Internet.
73,
Mark N5RFX
n5rfx
03-21-2007, 06:57 PM
Here are some 1/3 octave displays (http://home.roadrunner.com/~mdmiller7/male_female.jpg) of a male voice (Ted Koppel) left, and a female voice (NPR Female voice used in this thread) right. These were taken from the Internet and have not been processed since that time.
73,
Mark N5RFX
n5rfx
03-22-2007, 01:46 AM
One last post on this thread. I found a microphone and recorded 3 samples.
1st 200 - 3000Hz
2nd 200 - 4500 Hz
3rd wide open 20 - 16K (mic specs)
N5RFX audio file (http://home.roadrunner.com/~mdmiller7/n5rfx_compare.mp3)
73,
Mark N5RFX
Quote[/b] ]Here are some 1/3 octave displays of a male voice (Ted Koppel) left, and a female voice (NPR Female voice used in this thread) right. These were taken from the Internet and have not been processed since that time.
I can sure see why preemphasis is used when receiver design parameters limit highs to begin with. The energy in the upper 25% of the intelligibility range (3 kHz and up) is well below that of the lower frequencies. Thanks Mark!
So I popped a cassette tape in my truck's deck the other day... #Yeah, it's been a while since I've listened to cassette tapes, but I was in the mood for some stuff I only had on cassette... #Anyway, it occurred to me that Dolby NR is pretty much what we've been talking about here with the preemphasis/deemphasis. #I never really thought about it before, so I thought I'd drop one more post here for the heck of it. #It's amazing how all this stuff is connected in some way.
http://www.qrz.com/iB_html/non-cgi/emoticons/biggrin.gif
Joe
WA0LYK
03-22-2007, 11:14 PM
You know what, I had forgotten about Dolby NR on my old cassettes. There were times I would turn it off so that the highs came through better, drove the wife crazy though. With some hearing loss, ambient car noise (wind, motor, rear end, etc.), and poor speakers it seemed to make the music sound better, at least to me.
You are correct though, same kind adjustments to make it more intelligible and/or give it more fidelity.
Jim
WA0LYK
Yep, we've (finally) entered the Dolby age in amateur radio. I'm excited to see what comes next. I hope digital voice can be as good as SSB is right now. If we had the money, we could rent space on low earth satellites and use a digital voice protocol to communicate with each other -- but that smacks of repeater use to me -- and it doesn't give me the thrill of an old fashioned HF contact, which is still like throwing a line in the water and wondering what you might haul in.
I started in the hobby 31 years ago, listening to AM in the window on 75 meters, and the fidelity was always what I missed in my SSB contacts. Now it's possible to get the same fidelity on SSB using half the space. Ham radio just keeps getting better!
Thanks for contributing to an excellent thread. The research, the experiments, the dialogue and exchange of ideas -- even the arguments, were all very interesting and informative.
Til next time,
K3VR
n5rfx
03-23-2007, 02:29 AM
How Dolby Noise Reduction Works (http://en.wikipedia.org/wiki/Dolby_noise_reduction_system)
Making Cassetts Sound Better (http://www.dolby.com/assets/pdf/tech_library/212_Dolby_B,_C_and_S_Noise_Reduction_Systems.pdf)
I was always a dBx guy.
73,
Mark N5RFX
ke4pjw
03-23-2007, 03:00 AM
Quote[/b] (n5rfx @ Mar. 22 2007,13:29)]How Dolby Noise Reduction Works (http://en.wikipedia.org/wiki/Dolby_noise_reduction_system)
Making Cassetts Sound Better (http://www.dolby.com/assets/pdf/tech_library/212_Dolby_B,_C_and_S_Noise_Reduction_Systems.pdf)
I was always a dBx guy.
73,
Mark N5RFX
Ditto. The S/N and frequency response of dbx Type II was *FAR* superior to any consumer version of Dolby.
Quote[/b] (n5rfx @ Mar. 22 2007,20:29)]How Dolby Noise Reduction Works (http://en.wikipedia.org/wiki/Dolby_noise_reduction_system)
Making Cassetts Sound Better (http://www.dolby.com/assets/pdf/tech_library/212_Dolby_B,_C_and_S_Noise_Reduction_Systems.pdf)
I was always a dBx guy.
73,
Mark N5RFX
LMAO... Me too. In fact, I bought a second double cassette deck a long time ago with dbx because I was afraid I wouldn't be able to listen to all those dbx NR recorded tapes I have (and have them sound right) if my original died.
Funny thing is I came back to this thread to look up something, forgetting my Dolby NR post. This turned out to be a pretty enlightening thread, if you toss all the bozo comments (on either side of the topic).
Joe
k4kyv
03-28-2007, 11:24 PM
Quote[/b] (N3JI @ Mar. 19 2007,17:46)]BTW, I'm chopping at ~3400 for two reasons: One is I don't think too many folks listen wide enough even with the IF shifted to hear beyond that, and the second is simply to keep the QRM down. So I could go almost another full kHz, but I'm sure my neighbors on the crowded band wouldn't care for it.
When the band is congested I chop at 3400~ using a very sharp cut-off brick wall low pass filter. The low frequency response is flat to below 40~.
Feeding flat-response audio into the speech amplifier sounds somewhat dull and muffled, but by adding pre-emphasis, starting at 800~ and gradually rising to about 9 dB at 2000~, and then levelling out to a flat response plateau up to the cutoff frequency of the filter, the signal sounds clear and crisp, with fairly natural sounding quality. I often get "broadcast quality" reports even with the 3400~ cutoff.
Quote[/b] (k4kyv @ Mar. 28 2007,17:24)]Quote[/b] (N3JI @ Mar. 19 2007,17:46)]BTW, I'm chopping at ~3400 for two reasons: One is I don't think too many folks listen wide enough even with the IF shifted to hear beyond that, and the second is simply to keep the QRM down. So I could go almost another full kHz, but I'm sure my neighbors on the crowded band wouldn't care for it.
When the band is congested I chop at 3400~ using a very sharp cut-off brick wall low pass filter. The low frequency response is flat to below 40~.
Feeding flat-response audio into the speech amplifier sounds somewhat dull and muffled, but by adding pre-emphasis, starting at 800~ and gradually rising to about 9 dB at 2000~, and then levelling out to a flat response plateau up to the cutoff frequency of the filter, the signal sounds clear and crisp, with fairly natural sounding quality. I often get "broadcast quality" reports even with the 3400~ cutoff.
I'm sure it does sound great that way. However, my reason for cutting the low end is to focus as much power as possible in the range of speech that adds to intelligibility and punch-through. In other words, a full 1500W in the 350-3400 range instead of a lot of that 1500W in the 60-300 range...
Joe, N3JI
KC9JIQ
03-29-2007, 12:47 AM
Quote[/b] ]Dolby S-type encoding reduces the dynamic range (quieter sounds are brought up in level, while loud
ones are left alone). In noisy surroundings, such as cars, trains, or planes, it is no longer necessary to
keep turning the volume up and down as the level of the music changes.
WTF http://www.qrz.com/iB_html/non-cgi/emoticons/rock.gif you mean the old tapes you had to turn up and down the volume all the time?
Some of you oldtimers, explain this to me
ab0wr
03-29-2007, 04:11 AM
Quote[/b] (KC9JIQ @ Mar. 28 2007,17:47)]Quote[/b] ]Dolby S-type encoding reduces the dynamic range (quieter sounds are brought up in level, while loud
ones are left alone). In noisy surroundings, such as cars, trains, or planes, it is no longer necessary to
keep turning the volume up and down as the level of the music changes.
WTF #http://www.qrz.com/iB_html/non-cgi/emoticons/rock.gif you mean the old tapes you had to turn up and down the volume all the time?
Some of you oldtimers, explain this to me
No, this was just a way to get the quietest parts of a the audio above the built-in tape "hiss".
Tape has a limited dynamic range so you can't just amplify passages with a large dynamic range so the quietest parts are above the hiss level. You have to be selective in what you do.
All kinds of compress/decompress and preemphasis/deemphasis schemes evolved to do this.
Think of it kind of like the midnight position on your home theater receiver that lets you hear the whispers on TV without the loud passages waking up your wife! The only difference is that these schemes had both an encoder in the recorder and a decoder in the playback unit.
tim ab0wr
Well, After being told, here in this thread, that the problem I was having with ESSB signals, was because of my TS-2000 rig was not good enough, I went out and purchased a brand new FT-2000 and a Nice used TS-870S.
Now , I am proud to say, I am telling some ESSB users that they do in fact have "HORRIBLE " signals and the good news is, they are correcting them.
So, all this time, I was correct when I stated that there are many Horrible ESSB stations out there.
Boy, I sure do feel better that I spent $ 3500.00 to prove I was right all along!
Of course, the fact that I now have 3 more great rigs doesn't hurt either. hi hi...
It's good to be me ! http://www.qrz.com/iB_html/non-cgi/emoticons/tounge.gif http://www.qrz.com/iB_html/non-cgi/emoticons/tounge.gif http://www.qrz.com/iB_html/non-cgi/emoticons/tounge.gif
Gordon
Quote[/b] (KC9JIQ @ Mar. 29 2007,01:47)]Quote[/b] ]Dolby S-type encoding reduces the dynamic range (quieter sounds are brought up in level, while loud
ones are left alone). In noisy surroundings, such as cars, trains, or planes, it is no longer necessary to
keep turning the volume up and down as the level of the music changes.
WTF http://www.qrz.com/iB_html/non-cgi/emoticons/rock.gif you mean the old tapes you had to turn up and down the volume all the time?
Some of you oldtimers, explain this to me
WTF ??
Does that stand for "World Trade Forum?
I know it can't mean cussing of some sort......
Gordon
Gordon, were they really ESSB stations (more than 3,000 Hz wide) or were they guys with narrower bandwidth boosting bass and treble to obnoxious levels?
WTF -- world tree foundation?
Quote[/b] (K3VR @ April 02 2007,06:49)]Gordon, were they really ESSB stations (more than 3,000 Hz wide) or were they guys with narrower bandwidth boosting bass and treble to obnoxious levels?
WTF -- world tree foundation?
Some of BOTH!
I did not know that WTF stood for world tree foundation? ...
Cool !!
73,
Gordon