Speech Transmission Index - The no B.S. Guide for Fire Techs

In Part 1 we demystified 100 volt line.

Now, let's look at STI.

OK, stop sniggering - my colleague Claire is the worst. Every time I start talking about STI I hear a chuckle from her desk, so stop thinking rude thoughts, it’s not big or clever.

The Speech Transmission Index is built on over 100 years of acoustic and sound engineering research combined with a clever twist on classifying optical systems. It involves complex maths and complicated theories some of which are not fully formed whilst others are coming under closer scrutiny.

But fear not, as STI or the STIPA measurement has condensed all that into a 15 second test. A test you can do with a handheld meter, even your smartphone. That’s a triumph considering the full STI needs 98 ten second tests. We're talking 16 minutes to get just one result in one location.

Now to be serious for a moment; this mystery conceals booby traps that lay ready to blow big holes in your profits. So let’s dilly dally no more…

The mystery starts with what STI stands for?

Speech Transmission Index rates the amount a signal changes from the start to the end of its journey.

We’re interested in speech, not music, so the test signal has the main elements of human speech.

The STIPA test signal is based on male speech. Why? Because female speech is said to be more intelligible, using the male signal means it can only improve.

The test signal needs to take the same journey a real announcement would take: that's from the announcer’s lips to the listener’s earlobe

Transmission in STI means journey.

Three important facts:

1. The journey should start at the announcer’s lips, although it can be an input to the equipment.
2. It doesn't care about the announcer’s speaking skills or the listeners’ hearing acuity
3. The test assumes the listener has one ear – so actual intelligibility should be better.

The test compares the original test signal to that in the listening area. Index means rating.

• 1.00 STI is best - and impossible
• 0.00 STI is a disaster – actually more likely your meter is not working, and
• All STI ratings should be 0.50 or above, although that is tricky to achieve.

STIPA is a way to get an STI result, so don't make my mistake by saying 0.6 STIPA the correct format is 0.60 STI.

What does Speech Transmission Index measure?

I'm using African animal names for this. No reason, except my childish humour. I'm imagining you mouthing baboon, hippopotamus, gazelle and lion and those around you thinking you've finally cracked.

First baboon...

When you say it some parts are louder than others.

‘Ba’ starts loud, then tails off, ‘boo’ does much the same, with the quieter ‘n’ to finish.

Here’s a graph of ‘Ba’, ‘boo’, ‘n’ based on their loudness, officially called intensity.

‘Ba’ and ‘Boo’ start similar but sound different. Those differences are for two reasons:

1. Contained in that shape are faster peaks and troughs and these are different for ‘ba’ and ‘boo’, and
2. ‘boo’ takes longer to say than ‘ba’ - duh.

The proper name for that loudness shape is the intensity envelope, and the proper name for those peaks and troughs is modulations.

Those faster modulations are not even the same speed, so STI groups them by their speed.

Which brings me to hippopotamus...

To say ‘hippopotamus’, your lips and tongue move more and at a faster rate compared to ‘baboon’: it has more faster modulations.

Imagine visiting an old church with a mate. You tell your mate, you will say two words to them. You walk ten metres away, turn and first say, “Hippopotamus” then a few seconds later, “Baboon.”

Once your mate has stopped laughing, they’ll tell you hippopotamus was harder to understand than baboon. That’s because an old church will be echoey and faster modulations don’t like echoey. I know the correct word is reverberation.

Reverberation Is Like Snow

Stay with me, it's not as mad as it seems.

Picture a place you know well, your garden, street or outside your office.

On a summer day everything is visible, you can see tiny ants running around to large structures like cars, garden tables and sheds.

Now when it snows detail is lost. With enough snow, kerbs and steps get hidden. Large structures, like cars, are still visible, but the windscreen wipers have been covered and other sharp details smoothed over.

The bigger structures (cars and sheds) are like lower pitched sounds and smaller things (ants, windscreen wipers) are like higher pitched sounds. As the snow builds up so more and more detail disappears, with major snowfall cars are buried.

The amount of snow is like the reverberation time. A little snow covers the finer details and shorter reverberation times affect higher pitched sounds. Equally as heavier snowfall buries more detail longer reverberation times bury lower and lower pitched sounds.

STI's simple but clever bit…

Once STI has grouped the word parts by their modulation speed, it then works out how those groups differ from the original perfect signal.

If the changes are only in the fastest modulation groups, then reverberation time must be quite short and the STI number will be better.

With longer reverberation, the groups with slower modulations get affected, lowering the STI number.

So STI measures the effect reverberation has on speech?

Yes, it does, but STI deals with noise too.

Noise is like a lake

The beautiful place below has snow sitting on mountain tops smoothing out the detail (like reverberation). The lake covers all the low lying land, for all we know that lake might be hiding an ancient city.

Importantly, above the waterline the lake has no impact on the details.

So whilst reverberation affects the frequency or pitch of the speech part, noise submerges all speech parts that are quieter no matter their frequency. Sound geeks call this the noise floor.

How STI calculates the damage noise causes intelligibility

STI measures the average sound level, like a dB meter would, sorry should say SPL meter.

Without noise the average sound level would include all the quiet bits. Flood the place with noise and those quieter bits get submerged. The average sound level is now raised telling STI the intelligibility won’t be so good.

STI summed up with orange snow, a baboon and noise

We've covered how the bits of this mystery work separately, here's what they do together and how STI handles it all.

First that ‘Baboon’ graph, but with reverberation added.

The dotted blue line is the original graph you saw earlier, the orange looks a bit like snow sitting on top don’t you think? Yes, alright, orange snow.

With reverberation the peaks are louder which can be quite useful the bad news is that those earlier sounds linger so the troughs are not so deep.

Now we’ll remove reverberation and replace it with noise.

Just as the lake covers everything low down, so the quieter bits of 'baboon' are lost in the noise.

Put them together and we get this...

You’ll notice that the noise is also louder. Well, if reverberation makes the speech louder, it’s going to do the same for noise.

Speech Transmission Index in two sentences - well almost

STI measures the depth between the peaks and troughs (speech modulations). The less that depth difference the lower will be your STI number.

In the maths world they call such comparisons a transfer function.

That gives us the term Modulation Transfer Function (or speech journey comparison) and is the basis of the Speech Transmission Index.

Here’s the missing part: it's important

It’s called the Upward Spread of Masking.

I wrote the following sentence and it confused me, but still can’t find a better way of putting it - sorry…

A lower pitched part of speech makes it harder to understand a higher pitched part of speech that follows.

None the wiser? Me too and I wrote it.

BS EN 60268-16:2011 (the STI standard no less) puts it this way...

"When a loud, low frequency sound is presented at the ear, it will always mask higher frequencies, possibly rendering them inaudible if the difference between their relative levels exceeds a given threshold."

More accurate than mine but still confusing.

Time for a couple more African animal names me thinks.

First Gazelle...

‘Ga’ is pitched lower than ‘zelle,’ therefore ‘Ga’ makes ‘zelle’ harder to understand and the louder you say gazelle the worse this problem gets.

Importantly, this is caused by the way our ears’ work. This means shouting louder or turning up the PAVA system volume level won’t help with intelligibility or your STI number.

Now Lion...

‘Li’ is pitched higher than ‘on’ so does not suffer with this problem.

That was all easier with animal names.

This masking effect probably explains why loudhailers are useless.

That's the end of the beginning

You’ll be pleased to know if you've read this far you know the important things about STI.

The next section is about how Public Address and Voice Alarm kit affects the Speech Transmission Index.

If you've read enough, I’d understand

But before you go, we’ve just launched our own impedance meter called Zircle. When you are working on PAVA and other 100 volt line systems it is the perfect meter for doing maintenance, commissioning and fault finding.

Buy Zircle today and you'll save 35% which is a smidgen over ￡78 saving. And its free delivery too.

Also, here's the link to the first article in the series - 100 volt line demystified

OK if you’re sticking around, get comfy and we’ll carry on.

How Public Address & Voice Alarm kit affects the Speech Transmission Index?

 EN54 and its affect on STI

EN54 has two parts covering Public Address and Voice Alarm equipment:

• Part 16 for the Voice Alarm kit
• Part 24 for Loudspeakers

Neither have much to do with Speech Transmission Index (read EN54:16 Clause 1. Scope). However, equipment conforming to EN54 should be built to a good standard, minimizing its impact on STI.

That means that you still need to do a good a job on the design, installation, connections and configuration of the system.

EN54 is a big area so we'll cover different parts in future articles and with information on our website - visit www.zircle.co.uk where we'll add more free support.

But you should know one thing about it ...

EN54 is for PAVA equipment used as a fire warning.

If the Voice Evacuation system is for other emergencies such as crowd control, bomb or chemical alert and not for fire evacuation, then you do not need to use EN54 approved equipment. It's in the Introduction of EN54:16.

Using Your PAVA Skills to Make - not break - The STI Score

Like these cars, how you engineer your Voice Alarm system can transform it but in which direction?

In my view the effects of a Public Address and Voice Alarm system on the STI number are still not fully formed.

I think Speech Transmission Index started out as a way to deal with the acoustics of a space in concert halls where no sound system was used. For that STI, STIPA even RASTI can quantify the acoustics well.

However, amplify musical instruments and voices and all sorts of distortions arise. For some of these, STI struggles.

Even BS EN 60268: Part 16 : 2011 admits there are limitations to STI and STIPA when PAVA kit is involved. Additionally, testing by clever sound engineers and acousticians has found discrepancies between STI and actual intelligibility.

That is less a criticism of STI and STIPA, and more about the importance of you understanding what you are dealing with.

So although low grade PAVA kit can impair your STI number; you can do a lot more damage or good with how you design, install and set-up the kit. What you do, affects noise, peak clipping and distortion, and aggravates reverberation.

We’ll do a quick prime on these in relation to STI, and I'll follow up with more detail in future articles.

Noise - the weed of the perfect (and drug free) PAVA garden

System noise might blend in with the room noise so have no impact. But hiss type noise could sit over the 2 KHz range, which is critical to speech intelligibility and your STI number.

Hiss is often caused by passing too little signal between two bits of kit and then turning up a volume (gain) control after that point.

The good news is if you can find the mistake and correct it, you’ll boost your STI number for little effort.

Open Verdict on Peak Clipping, STI and Intelligibility

Peak clipping slices off the loudest bits.

It happens when too much signal is passed from one piece of kit to the next – the opposite to the hiss issue above.

Although research suggests peak clipping is not as bad for actual intelligibility as the STI number suggests, it’s not good for the kit and bad practice, so fix it you must.

Have You Seen or Heard of Lemon Face Distortion?

When I hear a distorted sound system I pull a face like I've just sucked a lemon. And it's not just me, I've seen others do the same.

Officially the cause of distortion is any change to the original sound, so that's everything we've covered so far. But, there are two specific forms of PAVA kit distortion which make us pull a lemon face: harmonic and intermodulation distortion.

For Speech Transmission Index, this is a tricky one as these types of distortion can improve or worsen your STI number.

Earlier I mentioned that STI groups the modulations by their speed. So distortion could create modulations that replace real speech modulations in the same groups, lifting your STI number.

Alternatively, if the added modulations are in the wrong groups or it shifts real modulations into different groups your STI could drop.

To avoid gambling with your profits you need to check for this - don't worry, there's more on this in the free members' area.

Is Reverberation Impossible to Control?

It's the result of sound bouncing off surfaces, so you could ask the builder to carpet the walls.

Some loudspeakers are designed to limit or control the spread of sound, this is known as directivity.

Rarely do cheap loudspeakers give you directivity info, and even if you can find it, they can be difficult to decode.

Interpreting spec sheets for all sound equipment needs its own article and I’ll do one soon.

Two connected issues make reverberation worse:

1. Loudspeakers spread sound wide at low frequencies, narrowing as the frequency rises.
2. Speech starts as low as 125 Hz with bits vital to intelligibility at around 2 kHz and higher.

If you know the coverage at 2 kHz and layout your loudspeakers accordingly, then those wide spreading lower sounds will bounce about.

This is where a graphic (or parametric) equaliser can help. You would reduce the intensity (loudness) in those lower frequencies so they can’t bounce about as much.

This also helps with that pesky upward spread of masking issue.

The Obvious & Easy Reverberation Fix

Point the loudspeakers at the listeners

People absorb sound so if you point loudspeakers at them there'll be less sound to bounce around.

Obvious as this is, look around next time you are out. You’ll see loudspeakers pointing in the wrong direction.

Some loudspeakers point upwards, perhaps communicating with aliens. Then there’s those 3 metres up or higher, aimed straight ahead, for all those going to work on stilts I guess.

There are two other factors related to the PAVA kit that don’t fit into any other category.

Compression and Limiting

These help protect your kit from loud sounds. With compression also able to boost quieter sounds.

Thus, both reduce the height difference between the peaks and troughs of the speech parts.

If your system needs this type of protection, then it is right that STI measures it.

Oddly, even though STI frowns on reduced modulation depth, compression and limiting, don't seem to impair intelligibility much.

My view is that you want to find out at what point the compression or limiting starts to do its thing. Then, if possible set-up your test signal so that it doesn’t hit that level, or only goes into the compression / limiting zone a little.

If your PAVA system needs compression / limiting when used in normal circumstances, then you have a seriously under-powered system.

To fix a lack of power is expensive, but there are some tricks to cut that cost.

We’ll cover this in another article , but just know 3 dB more level needs twice the amplifier power. That means whatever you already spent, you will need to spend again all for a measly 3 dB

The Trouble with STI Prediction and "A RANT"

You have to use software to predict STI. But that is only as good as the information you enter and getting that information is never easy.

I do hear from creatures who inhabit cloud 9, that the standards make it clear you must be given this information. Whereas you, who can’t have your head in the clouds, know that is laughable.

However, you have eight PAVA bids to respond to this month. None has acoustic information and all have woeful PAVA specs demanding 0.50 STI. Unhelpfully, BS 5839 : Part 8 : 2013 in the commentary for Clause 32 suggests you should do the work the paid advisers failed to do - free!

You could spend 30 hours on each design trying to fathom it all out, but you’d have no life and soon no business.

Alternatively, pay someone to do the predictions, but they still need the information. Even with the information it would be expensive to do for every tender of which you may only win 1 in 4.

Why not tell your client their tender documents stink?

In my experience that will offend the poor mites, no matter how politely you say it. They'll then give the work to someone who doesn't ask awkward questions.

OK, rant over

How to protect yourself from the tragic, but common, PAVA specification.

Without the acoustical information on the various spaces, consider including the following in your tender response.

----------------------

In addition to the performance of the PAVA system, intelligibility is affected by where and what materials are used in the building/room/space and their design and shape.

Often information about one or more of these aspects is difficult to gather at the tender stage or has not been fully specified. This being the case for this project, we have made the following assumptions:

• Reverberation Time (RT60) is under 1 second for each octave band centred on 125 Hz to 8 KHz.
• There are no acoustically coupled spaces.
• Noise levels (during an emergency) across all octave bands between 125 Hz and 8 KHz are within 3 dB and no octave band exceeds 53 dB.
• The STIPA test signal will be used for verification and set to 76 dB A @ 1.5 m AFFL.
• The announcer’s microphone is in an area that during an emergency has less than 40 dB A of background noise, and a reverberation time of under 0.25 seconds for the critical male speech frequency range.
• In all areas, meeting the above criteria, 0.45 STI or better will be achieved at 60% of all test positions.
• Tests (to be done and paid for by others) confirming the above acoustical parameters for each space are to be provided to us prior to the commencement of the STI testing.
• STI verification costs will be quoted once details of your testing requirements are provided.
• If areas to be tested do not suit STIPA testing as detailed in BS EN 60268-16:2011, then alternative forms of intelligibility testing would be done and paid for by others.
• Where STI testing is not regarded as suitable, then all outstanding payments become immediately due and all outstanding monies will then be paid by the next payment cycle or as agreed at the start of the project.
• Should more acoustical information become available, we may be able to improve these predictions and the final system’s intelligibility performance.
• Where a STIPA test or other intelligibility test produces a result below 0.45 STI at less than 60% of all measurement points or the equivalent on the Common Intelligibility Scale then further acoustical treatment or noise control measures may be needed. The design, installation and testing to implement and verify this additional acoustical / noise control work would be at additional cost to the price tendered and borne by others.

----------------------

Change as you want.

However, this doesn't let you off the hook for doing a good job on the kit selection, design, installation and commissioning.

How To Test Using STI?

Only kidding, we're drifting away from explaining a mystery. We'll cover testing in separate articles.

I bet you're relieved we've finished. I'm sorry for the length, but impressed and pleased you persisted. This mystery can do grave damage to your business.

Are You Ready To Take Your Business and Knowledge Even Further?

The article you’ve just read covered the basics of understanding Speech Transmission Index ... but that’s only the beginning.

We’ve been building our support and training business known as ProAudium for over none years, helping fire installers and system integrators with PAVA projects and problems

We’ve been fortunate to enjoy a lot of success in that time, but it took a lot of work ... and we’ve made our share of mistakes along the way.

What if you could directly benefit from those years of experience and the many before – and avoid those mistakes?

What if you could easily keep up with the state-of-the-art PAVA skills, tools and trends?

What if you could grow your PAVA work without wasting money on useless manufacturer training.

We’ve got something for you. We’re calling it PAVAcentral and it’s the first independent website for fire installers and system integrators covering this income booster.

It will go live soon so Click here to add your email address we’ll then tell you immediately we go live.

I hope you register and I’ll be in touch soon.

And here is part one if you missed it - 100 Volt Line demystified