Tuesday, January 26, 2010


One of the unusual skills one has to learn when recording English listening materials for China's educational publishers is a self-awareness about the speed of your speech.

Speaking speed is strictly graded according to the English level of the intended audience.

Fortunately, most of the work we do is practice drills for the English listening element of the gaokao - the national universities admission exam, held each June. This is paced at a fairly normal 160-165wpm.

Materials for the final year of senior high are also paced at about that level, although - unless they are specifically for intensive gaokao practice - we usually take a little more care to enunciate each word precisely, to lengthen our pauses just slightly. Second year of senior high is about 150wpm, first year 140wpm. Middle school materials are generally in the 110-130 range.

Beginners' materials get as slow as 90-110wpm. Materials for very young learners can get as slow as 70 or 80wpm, occasionally even 60wpm - which is just painful to do, really exhausting for the vocal cords.

The only exception to this framework seems to be Guangdong province in the south, where - whether through their greater historical exposure to English speakers in Hong Kong and Guangzhou, or a natural quickfire garrulousness in the local culture and temperament, or just from a perverse desire to be a little bit different from the rest of China - they ramp up the speeds by 10-15wpm across the board. This means that for gaokao, you're getting up towards 180wpm - which is way faster than most native speakers normally speak; gabbling, in fact. (It's also really hard to do, because you don't have time to read ahead, and so make far more stumbles. We're never given any time to pre-read or rehearse these scripts; they're all sight-read, so you have to try to read a line or two ahead of what you're actually speaking in order to get a sense of the context, decide on appropriate phrasing and intonation, edit out any Chinglish mistakes, mentally prepare for any tongue-twister moments, and so on. This is the other great skill of this work, which some people never really master. Voice recording work is often derided by other expats, but to do it well actually requires a lot of concentration and experience.)

So, a couple of weeks ago, my partner and I are asked to record something for a middle school listening practice book. We know how fast this should be, from long experience. And I think the scripts are actually marked as 120wpm. They also play us a snippet of a recording of one of our American colleagues reading the same script, as an example (as they usually do at this particular studio): yes, that's about 120wpm.

So, we start off recording it at 120wpm. But almost immediately we are interrupted by the young girl representing the publishers - can we go a little faster? Well, OK; you're the boss; anything to oblige.

Five minutes later, the same thing - faster. And again, faster. What?? Is this for Guangdong? No, no, north China, regular speed. Is this wrongly marked as being for middle school, then? Is it actually for 1st year senior high? No, no, middle school.

We're baffled, irritated - but we do as we're told.

The publisher girl is just going crazy: 4, 5, 6 times she interrupts us to urge us to go even faster. By the end, we're rattling through it at a full-on, gaokao 160wpm.

And you know what? That bloody girl butts in yet again with a request to speed up - on the very last page of the goddamn script!!!

It was no great surprise when we received a call from the recording studio last week to tell us that the publisher wasn't happy with our recording; it was much too fast; could we come in to re-do it?

Well, yes...... IF we're going to get paid for re-doing it. Fortunately, they were prepared to do that. So, yesterday we went in to give it another crack.

The same girl was there supervising the recording for the publisher. After 5 minutes, she came into the booth to ask us to go faster!! Ahem. NO. You know what, lady - WE know our jobs, and YOU don't. Just shut the f*** up, and let us get on with it.

You'd think, wouldn't you, that having cost her employer several hundred renminbi with her previous screw-up she might have learnt a lesson. It seems not. Some of the people I have to work with are just inconceivably dim. And it vexes me a good deal.

Deep breaths.


JES said...

Even if reading aloud from a previously read section, I cannot imagine ever talking that fast. On the one hand, I can sort of imagine what's going on: if the quality of the speech can't be objectively measured, let's focus on quantity. Still... I found one page with audio samples of the same passage being read aloud at speeds from 140 to 180 wpm; there, the consultant/teacher/author (a speech pathologist) says that the optimum rate is 140-160. You're right: if I had to work with someone who typically spoke at 180 wpm, I'd work with them as little as possible -- way too exhausting!

For a couple of years I volunteered at a Recording for the Blind facility in Princeton. I had stars in my eyes when I signed up, imagining that I'd be reading aloud works of great fiction or poetry. Declaiming, you know?

In the event, no such luck. They had tons of volunteers to read fiction, poetry, general non-fiction; what they needed were subject-matter experts. I ended up reading technical-reference books and software manuals... Maybe I should post something about that over at my place sometime -- it was much harder than I'd ever have imagined, from having to read punctuation in program/HTML code, through distinguishing among various types of "callouts" (warnings vs. advice), to describing detailed screen captures. (Yes -- it didn't make a LOT of sense to be describing a screen capture taken at (say) 800x600 resolution, when a sight-impaired computer user would be using a monitor scaled at (say) 100x75... if using a monitor at all.) I often wondered if anyone at all could actually make use of those texts, or if they were being recorded "just in case."

Don Tai said...

Good for you to be recording English and sticking to your guns. I know it's painful but it is so incredibly useful to Chinese students. They do need a lot of help.

I still remember "Special English" from VOA. People actually wanted me to use that speed when talking to them. It was much easier to improve my spoken Chinese.

Froog said...

Fascinating link, JES. It had never occurred to me that there might be a demonstration of this online.

After some years of doing this work, I have developed a pretty good internal sense of speech speed now - but this would have been invaluable to me when I was first starting out.

I suspect Americans, by and large (outside of the rapidfire north-east, anyway) speak rather slower than we Brits - especially in the south, and especially in the Deep South. I always joke with Chinese students that the only possible reason for them all loving the movie Forrest Gump so much is that they like the fact that he talks so slowly (not really a 'joke': it's true!).

I think my natural pace has probably slowed down somewhat since I've been here - partly perhaps from exposure to all the Americans (and Canadians and Aussies - all slightly slow), but mostly from slowing down so often to accommodate Chinese interlocutors. I think I typically used to speak at about 160 or so (maybe even a little quicker, when I get very excited about something), but have probably dropped to under 150 now - and will frequently slow down to around 130 for Chinese English speakers.

170wpm is the level at which it starts sounding just a little too fast for comfort. I've known a few people who speak at 180wpm or higher, but they are very hard work.

Froog said...

Well, there's the darnedest thing! I only had time for a quick listen to those speech samples at the start of the week, and it was bugging me slightly that the woman seemed a little fast - but I was inclined to dismiss it as just a case of my not liking her voice (I don't think she enunciates quite as clearly as we try to in the studio: she doesn't leave much of a gap between words, even when she's going slowly).

However, I just checked the samples more carefully with a stopwatch, and she is significantly above the stated rate - nearly 10wpm faster (she reaches the marker point in well under a minute, and in one or two cases stumbles over a phrase and thus adds in some extra words).

I am reassured to discover that internal sense of speaking speed is better than a professional voice trainer's.