ok so I realized that we didn't really have much of an opportunity to

look over many spectrogram during class today and so I thought it would be helpful

if we went through an actual spectrogram exercise and if I talked to

through some of the ways in which I approach reading a spectrogram and in this case

thought we'd do an SPHR faculty edition so the spectrogram show you

me reading the name of one of the faculty members here in SPHR

I'm not gonna tell you which one it is you'll probably figured out before we get to the end but it'll be a

a mystery until then

so here's the spectrogram and you can see that at the beginning starting out from brand new spectrogram can

haunting because it can be kind of hard to figure out where to start so when I looked at this

I actually made this recording quite a while ago and I didn't remember

exactly what was represented here so I had to figure it out myself right from the

beginning today so what I did is I started out by thinking well what

first and the first thing that I noticed is we've gotta sound right here and it's got a

after rustic structure to it and those look like formant to me and so the first thing that I did

as I mapped out the formants and I looked at those two formants and I thought well i've got an f1

around five hundred Hertz i've got enough to it's right around twenty-four hundred Hertz that's a

early wide gap that suggests that it's a vowel in the front part of the vowel space but I

not sure exactly where it is that's ok that's all I need to get out of that particular

a part of the spectrogram moving on to the next thing one of the things I

I noticed his right next to that there's something that's also very distinctive here

and that

is an area where you see things that look like they're trying to be formants

but they aren't really getting to be very strong formats in the way that I think

about this is that those format sort of look like crappy formats and

happy format tend to be characteristic of particular kinds of consonant sounds

Sana it sounds in general

with nasals and laterals being the main ones that have this kind of

kind of crappy format you can see there is one format there are about five hundred Hertz and another one maybe

about twelve hundred twelve hundred and fifty Hertz are so

so that's a nice little landmark now the next

the next sound that we have again goes back to having formant and you can see that their

f1 down there around four hundred hurts here show and f2 up there

around twenty five hundred Hertz so when you compare the first vowel we saw to the second

all that we see what I noticed is f1 went down and f2 up

which means that we're going up higher in the front part of the vowel space for

this avowal here to this vowel here

moving forward

what we see is that the next sound has two parts to it the first part is silence

and the second part is a release of that silence followed by a little bit of noisy energy

it doesn't seem to have any vocal fold vibration going on down

as I looked through all of that energy of the release

I noticed that it's not all the same it's relatively light down here but if I look

right here I see a whole bunch of darkness right in that zone

now that zone starts at about fifteen hundred Hertz and it ends around twenty five

it hurts so sort of centered right around two thousand Hertz and that's the akuma

of the energy of that particular burst just from looking at it I

tell that it's a stop and by measuring that band of energy

between fifteen hundred and twenty five hundred Hertz I can tell that the center of that energy is right

around two thousand Hertz and that's characteristic

villar stop in most cases

moving forward the next thing we see is another little vowel but notice how

short this vowel is compared to these other two vowels that makes me think that maybe this is kind of 

used vowel perhaps this is an unstressed syllable and we see that we've got a form in

had around five hundred maybe a little over five hundred hurts another formant up at

seventeen hundred eighteen hundred Hertz are so those are kind of in the mid

rangers both formant f1 and formant f2

the next sound we see

is has widely distributed energy and what I noticed

darkest energy is all the way up here at the top and in fact input

it's up above four thousand Hertz is where we really get into the darkness

energy that distributes all the way through the spectrum but it's a notable difference

we up here at the top and when I see energy that centered around four thousand Hertz

in a noisy sound that's usually a good sign that what we're talking about

is some kind of an alveolar freak IV so I'm thinking that this is an alveolar freak IV

the next sound after that

we have again silence followed by a a sudden release so I'm thinking

this is a stop but this time I don't see that same kind of accumulation of energy that I

I hear there's no extra darkness right around here and if anything

a little extra darkness up here now that where you have that extra

centered in the higher frequencies of around four thousand Hertz that's more character

of an alveolar stop and you might notice that there is a consistency between the

energy of the release of an alveolar stop like

and the energy that's accumulated for the alveolar frigate IV something like

that's not a coincidence that's a direct result of the fact that they're both made at alveolar

places of articulation

now the next thing that we see after that stop

another vowel we've got a first formant down

you're maybe at about six hundred Hertz then we've got this second format that's kind of has a trajectory

me that starts up higher and ends down lower if we aim for a mid part of

that we don't we don't know where it's coming from or where it's going so it's hard to tell what the former

auditions are right now but we may be able to say that there's a midpoint

that energy somewhere around two thousand Hertz are so for f2

so I labeled them as may be six hundred Hertz in nineteen hundred Hertz that's

eastern estimate and that's good enough for right now now this next sound that we're going to look at here you all know

crappy formant, formant that look like they're trying to be a vowel but kind of faith

and that's characteristic of either a nasal or a lateral sound and again

you'll notice there is a format

download

sort of crappy looking formant the exact location of these doesn't give you a whole lot of

affirmation but knowing that they're not very good formant tells you that this is some kind

us on a run consonant

the next sound that we have

as I look at it I noticed a lot of similarity between this sound and that sound in that

there's a lot of noisy energy

and there's an accumulation of that noisy energy up here around the four thousand Hertz

and I noticed one thing that looks a little different between this sound

down here what I notice is there's a little bit of evidence that there might

be vocal fold vibration going on some of these striations seemed to be

coming through here and that looks a little bit different than the noise

accumulating down here this does not look like vocal fold vibration where this

like it might be vocal fold vibrations so now I am wondering if this isn't

those are my hypotheses the next thing that I'll look at here

is another vowel and notice how this is a nice short vowel you got two formant that you can identify

they may make you think of these two formant they're pretty similar and in fact it's that short

I will again again I may be thinking of this as another reduced vowel

the next sound

looks very noisy

maybe not too surprising at this point but you'll notice that the darkest area they accumulation

energy is up here in the four thousand Hertz range so it looks like this name is a name that might have

the following sound

what we see are these crappy formant again looking a little bit like this sound

a little bit like this sound so we're thinking that this is us honor an consonant

and no big surprise what's coming up next we've got a vowel very clear f1

very clear f2 there is a little bit of transition that you notice

what may be hard to tell exactly what the format is exactly what the format value is

but you can tell that it's a good strong format trajectory and then finally

there's this here

it's clearly noisy it's not a periodic sound it doesn't look like it's got

any kind of vocal fold vibration and the noise is distributed in a particular way

it's very even from the top of the frequency spectrum down to the lowest

hurt where we see much noise being represented there's no real accumulation of energy

and where we see that kind of wide distribution of a

the energy of noisy energy we tend to think of this as almost

really the idea of is not very likely because there just aren't any words in English

names that I'm aware of that end with the sound

so I'm thinking is either a

sound hear at the end of this word now whose name

that this is actually a sound I think it's a dealer stop and it's clearly a voiceless

of voiceless stop

and here I'm thinking that this might beat

here I'm thinking if this is

I'm thinking this might be because it's a voiced alveolar phr KHA tive and here's another

sound so i've got a few of the sounds and I can start thinking well whose name

corresponds to this so here's a little list of the different faculty members in SPHR

 department i've included Nelson ，Nelson who is

not currently ah

down in the basement of new burger but she was on the faculty when I made this spectrogram so she was

in the mix and as I looked through the things that I'm gonna think of I wanna I wanna

that begins with a vowel then has a sonor and consonant and then has another

towel and ideally they would be two vowels in the front part of the vowel space

with this one being just a little bit lower

and this one which would be a little bit higher but both in the front vowel space I would want some

it has a dealer stop

and as I looked through these names there's one name that really kind of stands

I'm not gonna say that name here the other thing I should say actually buh

for I sign off the other thing I should say is this here is actually a pretty good

signature the idea that this may be the last sound in the name

Africa tive we would see an accumulation of energy up here which we don't see so this is probably a

non sibling something like a Ora

turns out there's only one name that ends that way so in fact the answer

to this ends up being reasonably easy

but the process of going through an identifying every single sound

can be very helpful in preparing for doing the kind of spectrogram reading

natural likely have to do as a part of the exam that's coming up so what i'd like to do

yes

turn you

the spectrogram with no annotations on it

and just suggest that you go through from start to finish and see if you can work out

exactly which sound is going on at exactly which point in this spectrum am

and see if you come up with the same answer that I did

thanks very much for watching and enjoy your spectrogram