Inspired by someone else’s attempt at another forum, I decided to record my first utterances in Swedish. Unlike most people who are just learning to speak, my first word wasn’t “mamma”; it was “sillgrissla.”
The excerpt is from Wikipedia:
Sillgrissla (Uria aalge) är en medelstor havsfågel som tillhör familjen alkor. Båda könen är svarta eller mörkbruna på ryggen, huvudet och översidan av vingarna vilket kontrasterar mot den vita buken. Den häckar i kolonier utmed kusterna i de norra delarna av Holarktis.
Not bad - I can barely hear any hint of Russian (the first time through, I came to think of Middle Eastern accent (Turkish/Persian/Kurdish?)). However, it sounds a bit like the way Alexander Arguelles and Stuart Jay Raj speaks Swedish. Some odd stresses here and there. The Swedish “grave pitch accent” (regardless of dialect) seems to be hard to grasp for learners. Usually it’s somewhat exaggerated, and as a result, some words have the wrong stress. A change in pitch accent doesn’t really move the stress, it just “changes”. Words like ‘svarta’ and ‘ryggen’ both have the stress on the first syllable, but the “way” the sound travels into the second syllable is different. In my Gotlandic accent, the grave pitch is similar to the Mandarin 2nd tone+neutral tone, and the acute pitch is the 4th+neutral - learners of Swedish (as long as they manage to realize that there are two pitches) usually pronounce the grave pitch as 2nd+4th - almost always with an exaggerated stress on the second syllable.
Let’s have a look at the word “grissla” (which has the grave accent) - as soon as there’s a prefix before, THAT part takes the pitch, and the rest get the normal (acute) accent (=sill…GRISSla). While ‘över’ is acute and ‘sidan’ is grave, the first two syllables in ‘översidan’ take the grave pitch, and the rest is acute. These things happen all the time: ‘brun’ (neutral), ‘bruna’ (grave), ‘mörkbrun’ (grave, stress on the 2nd syllable), ‘mörkbruna’ (grave, still stress on the 2nd syllable which leads into the 3rd where ‘-bruna’ actually becomes acute).
Here’s a link to an article you might find useful:
Similar to that Chinese doesn’t sound “Chinese” until we get the tones right, Swedish doesn’t really sound “Swedish” if the pitches are off.
The article is definitely very useful, but also confusing. For example, I don’t quite understand why they can define Right Melody as “usually a rapidly rising tone /LH/.” Since this is an acute accent, it is, of course, always falling in monosyllabic words.
I’m currently doing some sonogram analysis of words with grave accent, and my findings suggest that at least in bisyllabic words, (tillhör, for instance) both syllables have a falling tone but different offsets and different intervals.
The sample that I’m working with breaks down like this:
till-
Starts at about 265 Hz (C4) and goes down to about 160 Hz (D#3)—a minor sixth.
-hör
Starts slightly higher—at about 270 Hz (C#4) and goes down to about 127 Hz (B2)—more than an octave. Then it rises again to about 170 Hz (E3)—by about a forth.
So, the main difference is in the range by which the first and the second syllable change—a minor sixth vs. more than an octave.
Now, looking at the word leder (acute accent), I see this:
le-
Starts at about 155 Hz (D#3) and goes up to 195 Hz (G3)—a major third.
-der
Starts at about 165 Hz (E3), goes down to 125 Hz (B2)—a fourth, then goes back up to 165 Hz (E3).
These are, of course, standalone utterances and may not be representative of fluid speech.
This is where it starts to go over my head (i.e. references to frequencies etc.).
@astamoore - Do you find this kind of study effective?
Or does this kind of analysis keep your interest level up (being a musician, and perhaps someone who’s interested in ‘linguistics’) so that you can continue learning the language?
If you analyse more words (preferably polysyllabic, and of both pitches), I believe you could actually model your own speech according the results, as long as the corpus is from the same speaker (or at least a speaker of the same accent).
The wiki article about “ordaccent” ( Ordaccent – Wikipedia ) brings up dialectal differences, and the book “Vår fonetiska geografi” (=“Our phonetic geography”) has a lot to say about all the phonemes as well as the two pitches in the major dialect areas (it lists five areas, while the wiki page only has three).
It was a while since I skimmed through Kjellin’s article. At first I thought it might be how the two pitches sound in his own accent (who knows?), but maybe he’s referring to the acute accent as “rising” since you have to raise it just a tiny bit before the “release”.Think of how you would jump down in a swimming pool - you have to jump (up!) to get down. I agree that it’s very confusing, but this is the only way I can think of right now.
Thanks for the link to the article in wiki. A very good summary. I’m trying to use the information in it to analyze speech with some success.
A native speaker was kind enough to record this excerpt for me, and I tried to emulate her intonations. It turned out to be much harder than I thought. Here are two more attempts:
Don’t worry, you’re not getting worse! Just “different”. The native-ness varies from word to word, and from clip to clip. It’s not that one clip generally was better than the other. By the way, I can’t guess the accent of the native speaker (other than that she probably doesn’t speak skånska, småländska, halländska, gotländska, dalmål, östgötska, västgötska or “norrländska”).
Since you’re a musician, I’m pretty sure you can imitate isolated words. Anything can happen when it comes to sentences - similar to what happens when we’re playing a bass line or melody, opposed to a shorter sequence of notes. Some patterns require a certain rhythm/speed to “sound right” and to “be easy to play”. It’s easy to focus on how the individual words sound, instead of how the sentences (should) sound.
Thanks again, Jeff. Yes, her accent is, in fact, västgötska. I wasn’t trying to copy the pronunciation of individual sounds, though (too early for me to try to pick up an accent yet); just tried to form my sentence intonations after her (more or less).
OK, so it was västgötska. Maybe I can hear it in a word here or there, but the sentence intonation didn’t give a clue. Are you familiar with Olle Kjellin’s “chorus method”?
Oh, this is interesting (Kjellin’s method) and almost exactly matches my own self-study approach. The key difference being, naturally, that I get no feedback, except when I post a prerecorded segment (like in this thread).
I first read about the chorus method over at the HTLAL forum (“How to learn any language”) some years ago where the Swedish member Maxb mentioned that it was his approach to internalize Mandarin prosody. When we’re on our own we must adapt the method (since there’s no teacher at hand, nor a group of fellow students). I believe Max filled a sheet of paper (A4) with sample sentences, and repeated them over and over again.
