h a l f b a k e r y"This may be bollocks, but it's lovely bollocks."
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Imagine my surprise when, while reading about “Ebonics” (of all things,) I discovered that even though my family has been accused by other black-folk of “talking white” for nearly three generations Both me, my mother and my grandmother still use subtle speech inflections (such as dropping the hard ‘pt’
or ‘ft’ in works like ‘kept’ left’) that probably date back to whatever language our African ancestors spoke some place in the Ivory Cost.
Often people subjected to “standard” English in school learn to stop using the obvious markers of regional or cultural dialects, but it seems possible (and likely) that small variations in phonemes and sentence structure might still be used to trace a person’s heritage… or perhaps just determine where they’ve been. (Ever come back from another country with a bit of the accent?)
That’s when I got the idea for the online dialect test. You listen to a word or sentence said in a various ways and chooses the one that sounds most like the way you’d say it. Then the program determines what types of dialect make up your manner of speech. You also get to read the histories of the speech forms.
Bushonics
http://archive.salo...ushonics/print.html
[DrCurry, Oct 04 2004, last modified Oct 06 2004]
The Jive server
http://ccwf.cc.utex...ctic/toys/jive.html
You have probably seen it before but just incase you haven't. [Gulherme, Oct 04 2004, last modified Oct 06 2004]
ValSpeak
http://www.80s.com/...tainment/ValleyURL/
Gag me with a spoon! It's like totally bitchin'! [angel, Oct 04 2004, last modified Oct 06 2004]
[link]
|
| |
I like it. I sort of listen for these things in people's speech and try to pick out the influences. |
|
| |
I wonder if these markers could be cataloged and qualified sufficiently to accomplish this? |
|
| |
Hehe, //subtitle speech inflections//. Sounds like a halfbaked idea in and of itself. (typo fixed later) |
|
| |
No speech recognition needed. You just choose the one you think sounds like you. Expert analysis might be available for a fee, but then, I think normal linguists do that now anyway. |
|
| |
//speech recognition// Yeah, sorry. I skimmed it too quickly and made that assumption. Previous annotation amended. |
|
| |
Given the humongous range spoken English, you'd need a twenty questions style approach to narrow down the choices. That might then miss some important subtleties. Like your principal accent is Brooklynese (G*d help you), with major Ebonic influences, but you have a Welsh lilt inherited from your paternal grandfather. |
|
| |
Tempted, very tempted, to suggest amiblackornot.com, but it turns out to be baked. (Actually, it turns out to be half-baked.) Especially since "talking white" is a gross over-simplification. I think even your most pointed critics could tell the difference between, say, waugsqueke and myself. (Though blissmiss thinks I sound like UnaBubba, so maybe not.) |
|
| |
Finally, I will note that New York standard now sounds "normal" to me (ugh, again), and I have lost the ability to easily tell most British accents apart, even though I still have an English accent (or so I am told). |
|
| |
Where's Professor Higgins when you need him? |
|
| |
This idea is great (+). However, I find that you may need to incorporate some kind of speech recognition into it because though you might implement some kind of system like what DrCurry suggested, I think that no system where the computer talks to you first will work. People, in general, tend to pick up other's speech patterns or speak like those around them. Often, you might assume different inflections with your friends than with strangers or relatives. By having the computer ask you what you sound like by playing samples, the user may not be able to identify what his "normal" patterns may be because he might be apt to adapt to one of the patterns that he heard and perhaps likes. I was fortunate enough to attend a linguistic seminar where the professor first asked, "Say to yourself the word C-O-T then C-A-U-G-H-T" Thus, the audience would have no bias when saying the words and the professor would then identify the speaker's linguistic origins. However, in doing so, the computer would also need to have speech recognition capabilities. |
|
| |
wan-fu has a point; people may subconsciously choose the dialect they *want* to sounds like. |
|
| |
Like, of course, I speak exactly like Mr. Howell on Gilligan's Island did. What do they call that, Mid-Atlantic lockjaw or something? |
|
| |
I met a fellow when in high school, who himself was also high school age. He had an amazing skill for someone his age. He would say, "Recite the alphabet." After you complied, he could tell you what part of the United States your family came from. Even though I had moved from New Jersey to Georgia some 5 years before I met him (and therefore my accent had mutated significantly), he correctly identified the region of my birth. |
|
| |
I wonder if this process could be automated? |
|
| |
Dimandja: okay, I'm from Africa, explain my accent in detail! |
|
| |
Breaking this down cladisticwise for a "twenty questions" might be difficult. I think at each decision point the listener could probably be expected to remember and choose from only three or four alternatives maximum. What if the alternatives to choose from are: Georgia, Maine and Minnesota but the speaker is from Texas? Much less Britain. |
|
| |
This is why I think the voice recognition needs to be an integral part. The speaker would be shown a phrase and asked to read it. The computer would riffle thru dozens of stored profiles, chosing the one where your words were interpreted as the printed phrase. |
|
| |
This leaves out the "why" part of the idea. Watching Gangs of New York, I really admired Daniel Day Lewis doing a protoBrooklyn accent (none of the others could pull off anything comparable). I wondered how accents evolved. Chicago has a distinctive accent, but not as distinctive as New York. Is it evolving in that direction? What drives the development of regional accents? [Futurebird] suggests it is the remote influence of other foreign languages (african in this case) but in that case the Australians should not have developed such a distinctive accent. |
|
| |
Bungston: choices would not be states, per se, but how do you pronounce "class," do you say "got" or "gotten," etc., a few quick and broad brush determinants, before getting down to localized specifics such as how do you pronounce "Long Island," or signifiers that your maternal grandmother came from Ebonia. |
|
| |
According to a linguist I once heard on the radio, accents do have a tree structure, with cross-fertilization, so this does seem practical. (Modern Queen's English, for example, is pretty much a straight cross of Cockney and Dorset or Somerset, resulting from the influx of sea captains into the court of Queen Elizabeth I. Southern American is that accent slowed down to a drawl. I forget the rest of it.) |
|
| |
Bliss: Spanish is only enjoyable for non native speakers exactly because of that: it sounds nice to you. But believe me, it is a REALLY complicated language and, to me, any other tongue would be 10 times more interesting to speak. Spanish has too many words and conjugations. Not that I had a hard time learning it since i was born and raised in Mexico, but "dreaming" in spanish, as you so much wish, would mean you are familiar enough with the language, wich would mean that you would have to live either in a third world country like mine or in spain. And living in a third world country is not fun. |
|
| |
Dreaming in spanish can then become a nightmare!!! |
|
| |
I speak differently depending on my mood...(or which personality has just taken over my brain... :P), so would i just frazzle that thing or what? |
|
| |
Presumably the device could pinpoint precisely which soap character you are channeling. |
|
| |
I actively try to impersonate other accents as best as I can, and would do my best to try and fool the thing. The Russians I met last year thought I was raised in Kiev, until I flipped back into Americanese. |
|
| |
I've basically determined that UK as opposed to US English is a bit more efficient in vocal speed, simply because of the US tendency to overexert in facial movement. If I ever get a chance to visit a UK baker's convention, I'll have to see how close I come. |
|
| |
croissant, futurebird. this is fabulous. |
|
| |
with regards to the question of whether the system should play samples for people or emply voice recognition, the latter will produce infinitely better results. |
|
| |
in the linguistic work i've done (in university, not "professionally"), we were taught that people's accents are very succeptable (sp?) to suggestion. for instance, a person from new jersey says the word 'drawer' (as in where you keep your socks) in the exact same way that they say the word 'draw' (as in, draw yer six-shooters or draw a picture). |
|
| |
however if i'm quizzing a person on their accent and i ask, "how do you pronounce the word 'drawer'", they will inevitably pronounce the word with a combination of their own natural inflection and the inflection that i used to speak to them. |
|
| |
[bungston] spot on with the gangs of new york reference. though not really proto-brooklyn as much as, well, proto-lower-east-side-working-class-which-itself-melded-with-brooklyn... it was the only (and i do mean only) good thing about the whole film. |
|
| |
accent tracking is a fascinating topic to me because a person can involuntarily change their accent in a few months, even faster if they work at it, but a region's accent remains for generations. |
|
| |
I'm reading Jonathan Raban's "Badland" at the moment, about the settling of Montana. He talks of a rancher whose family had lived there for three generations, but he still spoke with a Norfolk (regional English) accent - the ranch was so big and his family met so few other people that they'd never picked up any other accent.
Children learn their accents predominately from their peers rather than their parents, so this system should pick up where you went to school as well. |
|
| |
I read something of some relevance last year: It seems, according to the writer of the article, that accents are developed from childhood-play-speak. The root of most creole languages is from kids speaking different languages playing together and mixing tounges. |
|
| |
[rods] lol. the concept of beard history warrants a thread all its own. cheers for the laugh. now, perhaps a web-based beard interpreter? input a beard and the software gives you its historical significance, periods of popularity, etc... then you'd need a beard scanner, of course... hmmm... |
|
| |
[Rods] - likewise, my daughter went to preschool in Palo Alto, California where she learnt a very definite Californian accent (particularly on words like "jelly", which uses a soft "ll" sound which UK English doesn't have). When she started school in London her teachers were very amused. After a few months though the accent had gone. |
|
| |
Likewise with me. When I got home from two years studenting in Stoke-on-Trent I realised that I had developed a pronounced midlands twang to my normal home counties, huntin' and shootin' accent.
That's why I'm convinced that some sort of speech-recognition element would need to be incorporated into this idea. A lot of the time, people simply don't realise what they really sound like when they speak.
This also goes beyond just regional accent. I always thought that I had quite a lively, expressive voice until I had a go at the voice substitution software on the Shrek DVD. For someone who delivers a small number of training courses each year, it came as quite a shock to learn how flat and lifeless my voice sounded. It was nothing like what I hear in my head when I speak. |
|
| |
Does anyone actually sound, to themselves - from inside their head, the same as they sound to other people? I've been kicking around an idea that is sort of related to that thought. |
|
| |
Military brats tend to lose regional accents while living overseas. You can generally tell the new kids on the base - they still have an accent and it's often pretty easy to guess where they come from. It would be interesting to use [futurebird]'s test to measure how rapidly accents blend. + |
|
| |
// Does anyone actually sound, to themselves - from inside their head, the same as they sound to other people? // |
|
| |
No, and most people much prefer their internal voice. The additional body mass the sound travels through adds a depth and resonance that simply isn't there in their external voice. |
|
| |
// A good example of this is the version of French spoken by the Quebecois. // |
|
| |
Acadian French is different still. In fact, there are at least four distinct groups of Acadian French and every one of them has a different dialect. In some places it's so strong that even non-Français speakers can hear the accent clearly. |
|
| |
In one part of southern Nova Scotia, it's more of a creole, really, combining English and French into one language. It's not uncommon to hear phrases like "drivez le car to le store". I actually heard someone say, while 'driving le car', 'flashez les lights on et off". The other Acadians make fun of this, of course. |
|
| |
Now going on six years living in the US, I've lost my Canadian accent more or less. Sometimes it betrays me, but more often than not it's because of a word usage, not a discernable accent. I'll never forget the first time I said the word "eavestroughing" to an American. |
|
| |
Futurebird's system--an audio database of regional pronunciations--could also be used in the other direction:
to train people in the nuances of dialects other than their own (for actors, public announcers, or people wanting to lose an accent/assimilate a new accent.) |
|
| |
[DrCurry]'s reported "English accent" may be noticeable only to Americans because it's different from what they are used to hearing; he may well seem, to UKers, to have an American accent, for the same reason.
My hometown has a recognizable accent, distinct from those of towns fifteen miles away (although this is becoming less the case); indeed some areas of the town, as small as 1/4 square mile, have their own accents. Although she cannot hear it, my mother has a noticeable "town" accent, and my father has a London accent. My accent is different from both, as are my brother's and my sister's, although we grew up together. My brother and my sister both lived in London for a time, but their accents are more of this town than is mine.
On my regular visits to Cornwall, I take about two days to pick up a (temporary) Cornish accent, which seems to be a hybrid of the several regional variations spoken by my friends there. |
|
| |
//most people much prefer their internal voice. The additional body mass the sound travels through adds a depth and resonance that simply isn't there in their external voice.// |
|
| |
Is there a device that can record your voice and change it so that it sounds like you think it does? It would be interesting to hear what other people think they sound like.
Afterthought: might make a good website. |
|
| |
I wonder what my accent would be? My parents are from Michigan, I was born in Pennsylvania, grew up in Illinois, and have lived most of my life in Colorado. |
|
| |
I have met several different people all born in the same town with different pronunciations of certain words, so a certain factor must be the parents. |
|
| |
[Don Quixote]: the American Army seems to have developed its own accent, which is similar to a Southern American accent spoken rapidly, with plenty of acronyms. |
|
| |
I wonder how a Colorado accent would be distictive? I believe it borrows heavily from California and Texas as well as the American Midwest (Ohio, Illinois, Kansas areas) which all blends together into a rather rural sounding variation on the Standard American accent, with hints of Spanish. |
|
| |
It is going to get tougher as the influence of TV invades regional accents. Here in the UK, it is quite noticeable, particularly amongst the young, how the influence of Australian soaps has overlayed accents with AQI (or high rising terminal) over the last 15-20 years. In many cases, it is more pronounced than native antipodeans - I hardly notice it in conversation with Aussie and Kiwi friends, but my niece's broad Cumbrian with a side order of AQI grates terribly.
[EDIT] After googling AQI, I've found some sources discredit the soap theory, and claim that some UK accents already exhibit high rising terminals. However, these regions are unlikely to have come to the attention of the mass of the UK population and caused the explosion of this inflection in the early to mid 1990s. Maybe it was "Beverly Hills 90210" that caused the problem. |
|
| |