[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Clearspeach speakers

Gunther Meisse wrote:

I am using the speaker within the Icom 910H and it will not win any awards
and is aimed at the ceiling. I want to put one on the wall pointed at ME.
Additionally, where can one purchase same?

Wayne replies:

I agree that intelligibility of a speaker improves when it's pointed
directly towards you.  That way the direct sound is proportionally louder
than the multiple reflections coming from the walls and ceiling.  (For an
extreme example, think of the difference between listening to someone talk
in a carpeted room vs. an office building stairwell)

Headphones provide still better intelligibility.  Then there are positively
NO reflections from walls/ceiling to muddy up the sound.  And you can
operate full-duplex without worrying about feedback.  Even in half-duplex
mode, headphones allow me to monitor my TX audio without risk of feedback.
I can easily tell by ear if I'm overdriving the ALC.

Looking at the web site for the Clearspeech speaker, it appears to be simply
"DSP noise reduction", similar to what is built in to many transceivers and
is available in numerous external DSP devices.  Some are much better than
others.  The DSP noise reduction in my FT-847 is essentially useless.  It
degrades the desired signal so much that there is no improvement in
intelligibility.  The algorithm is so bad that it mutes everything when no
voice is present for a few seconds, but doesn't adapt when the voice returns
(i.e., the desired voice signal is still significantly attenuated when it
returns).  After the FT-847 NR has been on for a while, it's as if you
turned the volume WAY down.

Several years ago I was very impressed with a borrowed W9GR DSP noise
reduction unit.  I recently bought a used Timewave DSP9+ unit for $50 at a
hamfest.  The DSP9+ noise reduction is far better than the FT-847 internal
noise reduction.  The notch is also much better.  Faster and deeper with
less degradation of the desired signal.  My only complaint about the
Timewave noise reduction is that it doesn't improve intelligibility of
people with mushy, bassy microphones (microphones that have poor
intelligibility even when the signal is strong).   The added distortion
offsets any benefit from reduced noise.  But the noise reduction is quite
remarkable when listening to people with good-sounding microphones.  In this
case, the added distortion doesn't really degrade intelligibility, but the
reduction in noise results in improved intelligibility and reduced fatigue.
I don't have to concentrate as hard to understand what people are saying.

Overall, I suspect that the effectiveness of DSP noise reduction is highly
voice-dependent, microphone-dependent, and listener-dependent.  The NR
algorithm adds more distortion to some voices than other voices.  The added
distortion is more tolerable if the transmitting microphone is bright and
articulate.  Some listeners are more bothered by distortion, while other
listeners are more bothered by noise.

A really good DSP speaker would have a microphone near your head, and a DSP
algorithm that actively cancels the delayed reflections that the microphone
picks up from the walls and ceiling.  Effectiveness of such a system would
be highly dependent on placement of the microphone.  High-end conference
room speakerphones use an opposite DSP algorithm to remove the speaker audio
from the microphones, resulting in full-duplex speaker phone operation.
They cost a LOT more than the Clearspeech speaker.

Gunther, I hope you will describe the overall performance of your new
Clearspeech speaker to amsat-bb.

Wayne Estes W9AE
Back in Beijing for a few weeks
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org