View Single Post
  #1   Report Post  
Old May 13th 05, 04:09 PM
Tom Holden
 
Posts: n/a
Default DSP audio filters: Timewave 599zx, MFJ 784B -- Useful for SWBC listening?

wrote in message
oups.com...

jamulc wrote:

What I'd like to know is if there's a DSP audio filter that really
makes any difference in making speech (on AM signals) more

intelligible
through DSP reduction of noise.

Best & 73,

jamulc


If greater speech intelligibility is what you want, then you should
consider one of the Clearspeech filters, especially the Clearspeech
"base unit". These aren't available new anymore, but you can find them
on ebay. In fact, one is up for auction right now:

http://cgi.ebay.com/ws/eBayISAPI.dll...sPageName=WDVW

(I have nothing to do with this auction, do not know the seller, etc.)
I've been using one of these for several months now and I love it. I
hardly ever listen to SSB without it. It's great at reducing "white
noise" and making weak signals more intelligible. (You will not want to
use it on AM, except perhaps under very unusual circumstances.)

Steve


NCT Group offers a demo software for processing short sound files
http://www.nctclearspeech.com/cspc.htm. NCT is the developer of the Clear
Speech algorithms used in the Am-Com unit listed above and in the newer Heil
Sound NCT Group Clear Speech Speaker listed at $210 at
http://www.heilsound.com/amateur/cle...ch_speaker.htm. The latter might
have more recent algorithms than the Am-Com.

The GAP http://www.gapantenna.com/hearit.htm HEAR IT Speaker for US$149 from
W4RT http://www.w4rt.com/ is the rebranded bhi NES10-2 Noise Eliminating
Extension Speaker
(£99.95)http://www.bhinstrumentation.co.uk/h...10-2_mkii.html. The SGC
ADSP2 DSP speaker http://sgcworld.com/shopsite_sc/store/html/page4.html is
priced at $129.95, claims 26 dB noise reduction versus bhi's claim of 35 dB
and appears to have auto notch filtering.

The noise reduction numbers may be misleading if the reduction algorithms
are different. Intelligible speech is the real objective - not noise
reduction per se - and all systems will have processing artefacts that
contribute their own unique distortions. The real story can only be found
out by subjective listening tests to the same material of wide variety.

Both bhi and SGC offer before and after sound files. It's possible to
process the bhi/SGC 'before' files though the ClearSpeech-PC software and
compare the results to their 'after' files. I tried that - it's pretty messy
but I thought the bhi/GAP came out on top. Their in-line module provides
more controls than their speaker so is more 'tunable'. A ham I know says he
prefers the bhi/GAP in-line module after having tried all the DSP boxes out
there.

Tom