Language Trainers Blog

I’ve written a lot about trying to listen to as much language as you can, whether it be in real life, or (even better) in a format that you can listen to more than once, and hopefully read along with.

I’m lucky to have a language teacher who has a very pleasant voice and manner of speaking, so when she records texts for me, it’s never a chore to listen to (although I do sometimes have to talk myself into listening to lessons rather than music or podcasts in English). However, I’ve heard a lot of people whose recording voices are almost painful to listen to, and I would struggle to make it through the few minutes it would take to listen to an article.

One particular example was an audiobook version of a book I’d already read and liked. The narrator had a grating accent and manner of speaking that emphasised all the wrong parts of the sentence and set my teeth on edge. I had to stop listening after less than a chapter.

I’ve found that I am more judgemental of speakers of my own language, though. I tend not to notice accents so much with foreign languages, and I’m not advanced enough yet to know if they are pronouncing things wrong. I haven’t found a foreign language speaker that I can’t listen to yet, which is probably a good thing!

Have you ever had to switch audio off because of the reader?

A voice mapping project by the British Library has collected thousands of voice samples from around the world. Speakers have recorded either six specific words, or Roger Hargreaves’s classic Mr Tickle*. Recordings are tagged with the place that the speaker grew up in, their age, and their gender. The map is based on the location the recording was made in (not the place the speaker comes from).

There are a lot of ways to look at this information, besides general interest and keeping a record of the way English is spoken todya. I read an interesting article about the fact that British accents aren’t being ‘Americanised’ (to be honest, I hadn’t heard that people were worried about the Americanisation of British accents or pronunciation, but apparently some people are). Based on the six words, controversy, garage, neither, scone, schedule, and attitude, investigations have shown that British people are definitely still pronouncing words differently from Americans. It’s interesting to note, though, that there is a tendency for British pronunciations to change, while American pronunciation remains the same.

The project also includes a lot of recordings from non-native speakers of English, which may help linguists to predict the direction of English pronunciation in the future.

The Voice Map project is on for a couple more days if you want to participate (you will need to have or register for an Audioboo account). If you’re interested in my recording, click here. If you want to explore the voice map, click here.

*Click here to learn more about why this book was chosen.

I’m sure it’s happened to most people before: you meet someone with a new, interesting, or strong accent, and somewhere in your conversation with them, you feel yourself start slipping into an (often awful) imitation of their speech. When you realise you’re doing it, you feel like a bit of an idiot, and you hope they haven’t noticed.

So, it turns out that accent imitation, besides being used for intentional and accidental mockery of others, can actually help you to understand the speaker’s accent. A study in the Netherlands has shown that if accented speech is repeated, it helps the listener to better understand it.

The participants in the study were first asked to listen to an unfamiliar accent of Dutch. To ensure that the accent was unfamiliar to everyone, the researchers made up a new accent by doing a consistent vowel sound change. After this, the participants were asked to listen to 100 sentences in the unfamiliar accent. Participants were asked to respond to the sentences in different ways: some repeated the sentence aloud while trying to mimic the new accent, some repeated it in their own accent, some transcribed the sentence, and some only listened. When the sentences were tested again, those that had repeated using the unfamiliar accent did far better than the other groups.

Unfortunately, in real life, even if you know this information it’s not very polite to talk to someone while trying to imitate their speech patterns. I suppose if you really wanted to, you could always use this study as justification.

Full article: Science Daily.

I’d heard that people are more likely to believe a person who speaks with the same accent as them (meaning people will be more likely to think that foreigners are untrustworthy*), and another study has shown some interesting results when comparing listener reactions to different accents.

The study, conducted at the University of Glasgow, used fMRI scans to analyse brain activity in Scottish listeners when they heard people speaking in American, English**, and Scottish accents. Participants listened to recordings of nine different female voices: three of each of American, English, and Scottish accents. The research found that brain activity was significantly different between listening to a native accent and a foreign one (even though all were speaking the same language).

“The pattern of neural activity differed strikingly in response to their own specific accent compared with other English accents,” Bestelmeyer said. “The initial results suggest that such vocal samples somehow reflect group membership or social identity, so that ‘in-group’ voices are processed differently from the ‘out-group.’”

My impression is that it takes a bit more work to understand a foreign accent, which would mean that people might unintentionally seek out people who speak in the same way, as it’s easier on the brain. However, I don’t think that this is necessarily related to in-group and out-group issues. It’s logical that a person finds it easier to understand the speech they have heard all their lives. What do you think?

Article: Science Daily.

*I wonder what kind of Google searches will be drawn in by this phrase.
**The article uses ‘British’, but I chose to use ‘English’, as that’s what I assume they meant (Scottish accents are also British accents).

I posted this on the Australian and New Zealand LT blog, with reference to my New Zealand accent, but I thought it was too funny not to post here as well. I’ve had some problems with voice recognition myself, but not as many as you’d think. Do any of you have strong regional accents that computers have trouble understanding?

In my last post I wrote about native speakers not knowing that language learners are not fluent.  I’ve experienced it many times when I have learned basic phrases in a language to be polite, and then have found myself apologising for not actually being able to speak the language.

So, is it beneficial for beginners to speak with a strong accent?  It gives the listener an instant clue that the speaker is not a native, and they can then (hopefully) modulate their language to suit the speaker’s level.  This saves you from having to explain that you are a student or that you don’t understand.

I’ve spoken to a few people about this, and opinions differ.  Speaking slowly or with a strong accent may benefit the learner in that the listener will modify their speech, but you can’t speak too haltingly, or ‘incorrectly’, or you may not be understood at all.  Most people don’t think that learners should actually try to speak with a strong accent (though some kind of accent is inevitable).

What do you think?

In the kind of news story that you wouldn’t even believe if it were in a Hollywood film, a woman from Devon started speaking in new accents after complaining of severe migraines.

She initially spoke in a Chinese accent for about a week, and then woke up speaking in a more Eastern-European sounding accent (apparent in the video below).  Unfortunately there’s no evidence of the Chinese accent; I would have been really interested to see what that sounded like.

Clearly she has had some kind of brain episode affecting her speech areas, as she makes some grammatical errors as well as the perceived accent change.  As she mentions in the video, she is unsure how long it will last for.  Apparently some people who suffer this kind of change recover after some time.

It’s nice that her mental capacity seems unchanged, although I think it would frustrate me no end to have people think I was a bit simple because of the way I spoke!

Source: Sky News.

I’ve talked about how people’s brains are continually trying to predict what’s going to come next, which is why we get misled by garden path sentences and sudden changes of direction.

A recent study of ERPs (Event-Related Potentials) by psychologist Jos J.A. Van Berkum from the Max Planck Institute in The Netherlands has shown that people’s brains show a specific spike (called the N400) when they hear a word that was unexpected or seems out of place.  Interestingly, the research also showed that it’s not just the words being spoken or read that matter, but other information as well.  As listeners, we also take into account clues from our perception of the speaker:

In addition to the words themselves, the person speaking them is a crucial component in understanding what is being said. Van Berkum also saw an N400 effect occurring very rapidly when the content of a statement being spoken did not match with the voice of the speaker e.g. “I have a large tattoo on my back” in an upper-class accent or “I like olives” in a young child’s voice. These findings suggest that the brain very quickly classifies someone based on what their voice sounds like and also makes use of social stereotypes to interpret the meaning of what is being said. Van Berkum speculates that “the linguistic brain seems much more ‘messy’ and opportunistic than originally believed, taking any partial cue that seems to bear on interpretation into account as soon as it can.”

The material used in the study was all in written or audio form, so there were no conclusions drawn about visual information about speakers.  I don’t think it would be a huge jump to assume that we also take visual clues about speakers into account, though.  It’s likely that we gather a lot of information about a speaker before they even open their mouth to talk. If they say something that doesn’t fit in with what we have assumed, it takes a moment to catch up again, much in the same way as it’s surprising to hear a tall, burly man speak with a high-pitched voice.

Full article from ScienceMode.

I’ve been working on an English pronunciation project recently, and we came up against the age-old (well, not really) question of American (AmE*) vs British (BrE*) English pronunciation. This particular project is required to use BrE, and the standard UK IPA symbols. I realised that if and when I need to create an AmE version, I will need to use a different set of phonetic symbols (notably for vowels), and teach some words in completely different ways. For example, AmE doesn’t really use the sound /ɔ:/ (the first syllable in the British ‘water’), but pronounces a whole lot more r’s than its British counterpart.

This led me to the question of choosing an accent. When you’re about to learn a new language, do you consider which dialect or accent you will be picking up, and will it matter in the future? Can the complete beginner even tell the difference? If you are learning English, should you pick a North American, European, or even Australasian accent? Apparently Canadian is the easiest to understand. If you’re learning French, do you want to speak like a Parisian or a Montrealer?

I know that when I was trying to improve my Cantonese, I preferred a Hong Kong accent to a country accent, and similarly I would rather have a Beijing or Shanghai accent than that of a small town that nobody’s ever heard of.

Of course this may sound like snobbery, but it basically boils down to increasing your chances of being understood. If you do your best to emulate a standard accent, people will be much more likely to understand you. I don’t think it matters if you choose American or British English, or American or European Spanish, as long as you try to learn a dialect that is commonly understood. If you go to a village an learn their particular accent and dialect, not only will you be an anomaly, but it will be much less likely that you will achieve the goal of learning a language – communication.

So if you’re making this sort of decision – just pick an accent and run with it.

*To prevent confusion, I am using American English to mean the standard or General American (GA) accent, and British English to mean a standard accent from England, sometimes referred to as BBC English or Received Pronunciation (RP). Not many people actually speak with either of these accents, but they are taken as a frame of reference. Please don’t leave me angry comments about how I should just call it English. English English is too confusing a term to use. As is Spanish Spanish.

I had the great pleasure of trying to place a collect call to Australia from the international airport in Bangkok (fairly impossible), and in the process I managed to stumble upon a curious detail in the automated voice system.  Using that particular phone company, I could place a collect call to the USA, Canada, or the United Kingdom.  The instructions were given for each country, and each set of instructions was given by a different recorded voice.  They had gone to the trouble of finding a Canadian voice, an American voice, and an English voice to repeat very similar information.

The fact that they had gone to the trouble of finding differently-accented voice talents to record their messages (or maybe just the one voice talent, instructed to do three different accents) was both intriguing and annoying.  Had they done any market research? Did it make people more likely to use their service if they heard a familiar accent on the other end of the line? Would people feel less stranded in a foreign country if they could relate better to the disembodied recorded voice?   I would like to see the data.

Meanwhile, I wish that time and effort could have been spent opening up the service to more countries.  Surely people pay enough money for collect calls to make it worthwhile?