From Newsgroup: sci.electronics.design
On 6/6/2026 7:55 AM, Martin Brown wrote:
On 05/06/2026 21:27, Don Y wrote:
In the POTS days, one could model a telephone channel as
roughly 3KHz BW (~300-3KHz) with the copper crapping out
at about 4KHz.
Digitizing to 8b at 8KHz and you were pretty much golden.
With cellular and the fact that legacy plant is still
in place (i.e., you have no real control over how the
signal is routed -- from second to second), is there
a better model?
Or, a technique to guesstimate the characteristics of
the "current" channel, dynamically?
You may be able to infer a rough idea of the channel frequency response by taking an FFT of a chunk digitised at 44kHz. But I'd be more inclined to low pass filter everything above 4kHz and use 8b @ 8kHz.
It should be good enough for the intended purpose of speech at that.
I typically am dealing with speech that is conveyed in open air
or via a relatively high bandwidth channel. I continually update my
models based on my knowledge of the speaker's identity AND THE
"unimposing" CHARACTERISTICS OF THE CHANNEL.
But, also have to handle the likely possibility of a telecom channel
for the same speaker.
This both distorts their perceived speech AND would bias the
model that has been built without those constraints.
If all you are doing is *recognition*, its not a problem. But,
for diarization and identification, it lowers their effectiveness.
Lack of sidetone and the latency of modern telco already leaves
you with a significant challenge!
As an "inverse filter" can't reconstruct characteristics that have
been discarded by a lower sampling frequency, I have two choices:
- run the trained models through an appropriate filter that
models the telco channel
- build a separate model for *just* the telco channel
But, if that channel can change from call to call... <shrug>
I *might* be able to use PRESUMED knowledge of the party's identity
to help characterize the current channel. But, that would defeat identification.
--- Synchronet 3.22a-Linux NewsLink 1.2