Instantaneous (analogue) compression of speech signals

In message , John Woodgate writes

It may be effective in conjunction with diode clipping - a small lamp will respond in (probably) some few milliseconds, so the initial harsh diode clipping would be short-lived and possibly less audible and less objectionable as a result. Speculation, naturally.

No, it was a poor joke. DSP may be over the top, but a PIC (or whatever) with ADC, look-up table and DAC would be pretty simple for a Lo-Fi implementation.

Cheers

--
Keith Wootten

It really does depend what you want. Generally, you can clip quite agressively, _provided_ you have an automatic gain adjustment before the clipping, and may actually improve intelligability if this is done right. The IC that used to be commonly used, was the Plessey SL6270 VOGAD (voice operated gain adjustment device), which massively reduced the dynamic range needed for speech. A search on this may find an equivalent. Are you trying to listen while speaking, or recording the sound. It is terribly difficult to 'judge' your own voice through such a circuit. Try using a text message recorded by somebody else, and playing it through the circuit. I might even be able to find a couple of the original Plessey IC's. We used them many years ago, as part of a speech digitisation system, and there may still be some in my cupboards somewhere!... Presumably you have lost most of the higher frequency response in your hearing. This has practically no effect on speech (you can filter everything above about 2.5Khz, and still understand speech perfectly), but would massively reduce the effect of treble 'boost' controls.

Best Wishes

I read in sci.electronics.design that Keith Wootten wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Tue, 4 Jan

2005:

It was a reasonable feed for a dead-pan response.

True, but anything like that raises EMC issues, which I don't want to get involved in. Analogue is much less hassle, if it works well enough.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

I read in sci.electronics.design that Roger Hamlett wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Tue,

4 Jan 2005:

Almost everything does. (;-)

Do you have any references for the increase in intelligibility? That's part of the bigger picture.

Yes, I have one of those in my assisted hearing device that I use in committee meetings. Google shows a number of NOS sources but I don't see a current equivalent.

That's what I'm using; the Canford Quick Check voice tracks, for example.

Yes, you've twigged it. I have lost a lot from 1 kHz up and that does cause problems with speech.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

I read in sci.electronics.design that Jim Thompson wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Tue, 4 Jan 2005:

This is helpful for theory but the devices are not now available, I think.

I have concerns about the 'direct' mode, which is clearly non-linear!

Not 'instantaneous' and data only in Japanese (:-(

Not accessible to me.

Not instantaneous; these use a rectifier and thus involve at least one time-constant.

Thanks for your help.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Digital stuff is just a passing fashion.

ANALOGUE INEVITABLY RULES!

You can't design a microprocessor without considering the inter-connections to be transmission lines with Zo, attenuation and phase delay.

I read in sci.electronics.design that Jim Thompson wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Tue, 4 Jan 2005:

I couldn't afford your professional services. And as yet I don't know what I want. I'm still at the breadboard stage. I have something that 'doesn't not work', but I want to know how far off optimum it is. And the sound quality is important but I can't hear well enough to assess it. I need 'ears-on' local assistance, and I have some colleagues visiting on Friday. Maybe they will listen for me.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Neither did I, and I wanted to confirm that up-front. (;-) Let's see how things pan out.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

We're practically twins in that regard, Jim, except mine was sudden, due to a skull fracture, and complete, profound, 100% loss in the left ear, with nothing left behind but the tinnitus. It makes it easy to sleep when laying on the right side, but I have concerns about smoke detectors, etc.

Tom

In article , Jim Thompson wrote: [...]

I like the homomorphic compressor, not because it is better in any way but because it is unusual.

Take the abs() of the signal but remember the sign.

Take the ln() of the abs()

High pass

do the exp()

Restore the sign.

--
--
kensmith@rahul.net   forging knowledge

In that case, use a voltage controlled amplifier and control the voltage by the filtering the output of a peak detector. I've used these circuits many times with excellent results.

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

For that you will need special equipment and or software. One of the simplest measurement is STI which stands for Speech Transmission Index. This will give you a number which says how good or poor the speech is conveyed. Bruel & Kjaer make equipment to do these sort of measurements.

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

I read in sci.electronics.design that Nico Coesel wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Wed, 5 Jan 2005:

This is not instantaneous. There is inevitably a time-constant associated with the rectifier filter capacitor. I, too, have used this technique, but it isn't what I want for the present project.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

I read in sci.electronics.design that Nico Coesel wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Wed, 5 Jan 2005:

I didn't mean measurements of intelligibility, I meant measurements of voltages. There are numerous pitfalls even in that apparently simple measurement.

The Bruel and Kjaer RASTI box is long gone. There are numerous STI and RASTI meters available now, but there is an increasing weight of problems and anomalies associated with STI. The study of this is another project in which I have a significant interest, and I am hosting a meeting on the subject on Friday.

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

I assume you are referring to speech clipping in order to increase the received volume level without exceeding the peak-power capacity of an AM or SSB transmitter.

The following is based on personal design experience albeit 25 years back.

It really does work.

There is a microphone gain control pot. Following the microphone amplifier is a simple 400 Hz to 3 KHz filter. The filter should be correctly terminated with Ro to minimise over-swing on sharp speech transients. Maximum available filter output volts being about 6 volts peak-to-peak.

There is then a 10K resistor followed by a pair of back-to-back small signal, high-gain transistors. The transistors, such as BC109's, are diode connected with base connected to collector. The transistors behave as clipping diodes with a much sharper than normal knee transition. The sharper the better!

The maximum output from the clipping circuit is plus or minus 0.6 volts. The amplifier following the clipping circuit should have a high input impedance,

100K or greater, so as not to interfere with clipping action.

Coupling capacitors following the clipper should be large enough to pass the lowest audio frequencies passed by the 400 Hz to 3 KHz filter without attenuation.

Then follows the remainder of the transmitter, either AM or SSB. A second gain control is needed after the clipping operation to set the transmitter drive level.

At a clipping level of 6 dB on speech peaks, speech is highly intelligible with hardly any distortion. Music is quite distinguishable. 6 dB is equivalent to a 4-times increase in transmitter power.

Clipping levels of 10 dB or more can be successfully used to improve received signal to noise and interference ratios.

===============================

There is a more complicated arrangement which marginally improves clipping performance but which slightly changes the tonal quality of speech. It may sound like a different person speaking in a different room. It requires a balanced modulator and an additional IF side-band crystal filter at 455 KHz, identical to the crystal filter in the main SSB transmitter.

The clipping is done at IF between the pair of crystal filters.

These excellent signal processing techniques went out fashion with homebrewing and when citizens' band came in.

--
Reg, G4FGQ

The circuit I used, uses 2 LM393 comparators (for positive and negative peaks). The outputs are open drain and can discharge a small timing capacitor quite fast. If you need a faster way, you need to go digital so you can use a feed-forward approach. A PIC processor and an 8kHz u-law serial codec should be enough.

--
Reply to nico@nctdevpuntnl (punt=.)
Bedrijven en winkels vindt U op www.adresboekje.nl

I read in sci.electronics.design that Reg Edwards wrote (in ) about 'Instantaneous (analogue) compression of speech signals', on Wed, 5 Jan 2005:

The application is not for radio, and I need 100 Hz to 5 kHz bandwidth. I have an active filter at the input, to condition mic and CD signals.

That's interesting. I have 10 kohm, and shunt 1N4148s with series resistors to *soften* the knees. I'll try hardening them, your way.

Yes, I have a TL072 buffer. 1 Tohm should be high enough!

--
Regards, John Woodgate, OOO - Own Opinions Only. 
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

If you could tolerate a time delay, you could do a very nice smooth AGC thing without the clipping problem that results from a fast-attack signal. A negative delay line (future predictor) would be handy here, too.

John

John,

The filter preceding the clipping circuit can be a high-pass (first) followed by a low-pass.

In a speech waveform, specially the male voice, most of the signal amplitude is contained in the lower frequencies, say 400 Hz and below. So the high-pass filter section tends to level the amplitude even before clipping.

Nearly all of the information is contained in the band 400 Hz to 3.3 KHz.

The clipper turns the large low frequency components which get through the filter into square waves. There's a peculiar effect. The human ear and brain partially succeeds in re-constituting the missing low-frequencies. Imagination? Dunno about dogs' ears.

The higher frequencies are distorted into much higher frequency odd harmonics. The function of the low-pass filter section is to minimise the effect.

A second low-pass filter can be inserted after (not immediately after) clipping to limit the RF bandwidth actually transmitted.

You can adjust filter frequencies to suit your own application. There is no interaction between the various cascaded circuit sections.

Oscilloscope patterns obtained with speech, music and sinewave inputs are very interesting while varying the two gain controls and listening on a loud speaker. Despite the sharp clipping action, the onset of clipping and distortion is quite soft.

An alternative clipping circuit is a cathode or emitter-coupled pair.

In my younger days I did a lot of testing and fault-locating on GPO music circuit transmission lines into BBC broadcasting and TV stations. Used headphones as the detector to balance impedance bridges up to 20 KHz. Found it possible to train one's ears to hear up to 20 KHz. Saved time changing from headphones to amplifier-plus-meter at 10 KHz where most people got stuck.

--
Reg, G4FGQ