"Citrus doesn't go with chocolate."
"Citrus doesn't go with chocolate."
People are always trying to foist those break-apart orange-chocolate orange things at me. Blech. I can sort of deal with lemon cream and chocolate, and if you must cover your candied orange peels in something other than sugar, I won't argue with you. But really, one of these things doesn't belong: citrus, coffee, mint, raspberry. Some things go with chocolate, and some things don't. Get over it. End of sermon.
[s], IPA 132
Lower-case S
So we have a nice, slow-rising amplitude fricative here. It's evident that it's strongest in the highest frequencies (since they are represented earlier and just get stronger as the lower frequencies come on, and centered above 4000 Hz somewhere. So this is a pretty classic [s].
[ɪ], IPA 319
Small Capital I
Tiny short vowel, but quite loud and high pitched. If you're tempted to ignore it due to its length, fine. But if you do, please tell me you labeled it as a barred-i. F1 is low (high vowel), f2 is highish (front vowel) but not clearly in [i] territory (usually 2100 or 2200 Hz or so for my voice). And not long enough to be historically long, so probably a) reduced or b) historically short (i.e. 'lax' whatever 'lax' means. Hence, [ɪ].
[t], IPA 103
Lower-case T
Plosive indicated by the gap in the spectrogram. No voicing bar so voiceless. Long VOT, so probably aspirated (but there's something else going on there). Transitions don't tell us much. So all in all, go with the most common member of the class. Honestly, I got nothing.
[ɹ̥], IPA 151 + 402
Turned R + Under-Ring
On the elsewise hand, we've got good evidence of a lowe F3, rising into the following vowel. This is why the transitions aren't tell you much about the plosive--any information in the transitions for the plosive are sort of being wiped out, or overriden, or hidden, by accommodating the low F3. So taking the lowest point of the F3, which is sort of in the middle of the aspiration/VOT, I called this thing voiceless, and the F3 being down around 1750 Hz really it can only be an [ɹ].
[ɨ], IPA 317
Barred I
Well, this is longer than the last one, but it's also moving hard and fast. The F1 is rising slightly, as is the F2, and of course the F3 is rising sharply back to neutral. So something that moves like that is mostly transitional, i.e. reduced, Treat it as such as move on.
[s], IPA 132
Lower-case S
A little stronger in the high frequencies and weaker in the lower, but still classically [s] shaped--broad band, centered in the high frequencies.
[t], IPA 103
Lower-case T
Another gap. This one a little more classically alveolar, what with the falling transitions in the following vowel, consistent with the rising transitions in the preceding vowel (although those are obscured by other factors). The falling F3 is really only something you should ever get with an alveolar.
[ʌ], IPA 314
Turned V
Vowel. So the F1 is rising slightly, but is storongest at the beginning at about 500 Hz. So this is middish and getting lower. F2 is just below 15000 hz and falling, so backish and getting backer. So this is not a middish to lower middish vowel, backish and backer. So there you go. Easy.
[z], IPA 133
Lower-case Z
Well, this is another alveolar sibilant, for the same reasons, but I thought this one was voiced. Now I'm not so sure. Oh well.
[ɨ], IPA 317
Barred I
And thank heavens this vowel is just too short to worry about. In fact, most people would ignore it. Certainly in transcriptions this sort of thing is regarded as a syllabic...
[n], IPA 116
Lower-case N
...nasal, but in practice, at least in American English, there's almost always a short (very, very short) period of open vowel before nasal closure in this sort of situation. So what we see here is a longish sound with a fairly strong voicing bar and very limited energy above it. It as fairly sharp 'eddges', which is another indicator of nasality (i.e. the sudden occurrence and offset of a side cavity). Hard to tell what's going on with place, since the nasal pole I usually expect in the F2 range (around 1000 Hz for labial [m] and around 1500 Hz for [n]) is not really supported by anything. So you might guess [n], just playing the odds, or you might just call it a "N" for nasal, and move on.
[t], IPA 103
Lower-case T
There's not a lot of evidence or non-nasal oral closure here, so ordinarily, I'd indicate this as just a release on the nasal, but since it appears to be voiceless, I didn't have a raised [t] symbol when I made up the solution spectrogram. So this is just the release (a T in Keating et al (1994)'s release notation, without an accompanying TCL closure). Itis very sharp/abrupt, and its noise is tilted to the high frequencies (i.e. [s] looking) rather than in the middle (F2/F3) frequencies, as we might expect for a velar, or the lower frequencies as for a labial.
[k], IPA 109
Lower-case K
On the other hand, this is clearly a plosive. It features F2/F3 pinch (i.e. they start close together and move apart into the following vowel) in the following transitions, and a nice mushy double burst, strongest in the F2/F3 region, absolutely classically a velar plosive. Voiceless, of course, but unaspirated.
[oʊ], IPA 307 + 321
Lower-case O + Upsilon
So except for the transitional blip at the beginning, the F1 of this looks pretty solidly "mid". So I think we're safe in calling this a mid vowel. The F2 seems to transition in a straight line downard from central to low (i.e. from central to back/round). This could just be transition, since when the F2 reaches it's minimum we also get a change in amplitude in the upper frequencies, but the fact that it's straight means a) that there's no F2 target in the vowel (unlikely) or that b) this is 'directed' or controlled or VISCy movement. So let's take it seriously and say this is centralish and moving backer and rounder. So pretty definitely a phonemic /o/.
[w], IPA 170
Lower Case W
On the other hand, straight F2 transitions like this, and low F2s with low energy above are pretty standard for [w].
[ə], IPA 322
Schwa
Short vowel that's mostly transition, and otherwise adjacent to a fairly longish, strongish, i.e. probably stressed vowel. So probably reduced. Moving on.
[θ], IPA 130
Theta
On the other hand, there's about 50 msec of 'something' approaching the 1200 msec mark. It isn't exactly gap-like, there's noise in the extremely low frequencies and some noise or clunks or something all through the spectrum. So given its clunkiness, this turns out to be a fricative. Not sibilant, which limits the choices.
[t], IPA 103
Lower-case T
Another gap. This one with nothing really useful on either side to give you much information. But phonotactics will limit your choices in a moment.
[ʃ], IPA 134
Esh
For instance, right now. We've got here a sibilant fricative. It's still mostly one broad band, but this time it is experiencing a little more filtering from the front cavities (we know this because sibilants really can only resonate through the front cavities, the slit effectively closing the back cavities off. There's energy in the F2 region, and substantially less energy below, which is suspicious, and the peak seems to be at 3000 Hz, instead of off the top of the spectrogram. So taken together, I say this is almost definitely an [ʃ].
[ɑ], IPA 305
Script A
At last, a vowel worth talking about. F1 and F2 sort of converge into a single wide band, at least at this resolution, but you can tell from the width that it has to be at least two formants. F1 and F2 converging (at about 1000 Hz) is a pretty typcical way of describing [ɑ], the lowest backest vowel most of us can produce without a lot of rounding.
[k], IPA 109
Lower-case K
And there's another plosive. If you take a good look at the preceding vowel, the F2 is rising and the F3 is falling, which is how we describe velar pinch. Well, we really like them to not only move toward each other but to actually make it close together, but one can't have everything.
[ɫ], IPA 209
Tilde L (Dark L)
Well, believed it or not there's a long VOT in here hiding a different segment. The Low F2 and slightly raised F3 indicate an American English dark /l/. At least partially devoiced due to the accompanying aspriation.
[ə], IPA 322
Schwa
Deceptively long vowel, due to final lengthening, but the F2 is all transitional.
[t], IPA 103
Lower-case T
The release isn't as sharp as I'd like, but the noise is vaguely [s]-shaped, so this is an alveolar plosive.
| Robert Hagiwara, Ph.D. Dept. of Linguistics University of Manitoba Winnipeg, Manitoba CANADA R3T 5V5 |
How To - Research - Courses To the Lab - To the Department - To the University |
 |