RS: Back with us from Los Angeles is Nina Weinstein, author of the English teaching book "Whaddaya Say? Guided Practice in Relaxed Speech." Last month she talked about reduced forms like wanna, gonna and hafta, the spoken versions of "want to," "going to" and "have to."AA: In the same way, "what do you" changes to "whaddaya" when a speaker is talking at a natural speed. Nina Weinstein says once you learn how speed affects words, the next step is to learn how it affects the pronunciation of sentences.

NINA WEINSTEIN: "There are languages that have choppy rhythms. For instance, German has a choppy rhythm, Vietnamese has a choppy rhythm. But English doesn't; English wants to be a smooth language. So the best way to smooth out a language, or the smoothest organization, is consonant vowel, which explains the first rule: If there is a final consonant sound followed by a vowel sound, the consonant sound will move over. This is automatic; people don't think about it.

"And we have a couple of examples that we could play if you want to, to show how the reduced forms and these pronunciation rules work together."WOMAN'S VOICE: "[with reduced forms only] I don't wanna spend a lot a money. I don't wanna spend a lot a money."NINA WEINSTEIN: "The written sentence is, 'I don't want to.' We learn that the reduced form for 'want to' is wanna, so that's how we get 'I don't wanna.' And then 'spend a lot of money' -- we learn that 'of' becomes 'a' if it's said at a natural speed. So we're left with 'I don't wanna spend a lot a money.' So now we apply the sentence rules.

"We want to know how the sentence will be pronounced. So we look at 'spend a' and we know that 'spend' is said with a final consonant sound followed by a vowel sound, 'a,' so instead of 'spend a,' we will get 'spenda': 'I don't wanna spenda.'

AA: "What's your other example there?"NINA WEINSTEIN: "Well, the other example in that sentence is 'a lot a.' So 'a lot a,' we have the 'o' sound before the 't' and the 'a' sound after, so we have a 't' surrounded by vowels. 'T' changes to 'd,' 'd' is a final consonant.

"Usually a final consonant sound just moves over to the next vowel sound, but we have a special case if it's a 't': we want to make sure to look for vowel sounds or listen for vowel sounds on either side. So we have 'spenda loda.' With 'lot,' 't' becomes 'd' and then it moves over to the 'a.' So it's happened two times in this sentence: 'I don't wanna spenda loda money.'"WOMAN'S VOICE: "[with reduced forms and sentence rules applied] I don't wanna spenda loda money."RS: That voice is from the practice materials with Nina Weinstein's book "Whaddaya Say?" The next rule she teaches is that when two of the same sound are next to each other, one is going to drop out.

WOMAN'S VOICE: "[with reduced forms only] Git em as soon as possible. Git em as soon as possible."RS: Said slowly that is "Get them as soon as possible."WOMAN'S VOICE: "Git em as soon as possible."NINA WEINSTEIN: "So 'as' ends with a 'z' or close to an 's' sound. 'Soon' begins with an 's' sound, so two of the same sound, one drops out. And, again, it's that smoothing of the rhythm of the language. If we stop to pronounce each sound, we're going to break the language into a piece. So instead of 'as soon,' it becomes 'asoon.'

"Now we apply the sentence rules and we look at 'git em.' We have a 't' sound. We said whenever there's a 't,' we want to know if there are vowel sounds on either side, which in this case there are. So 't' becomes 'd' and then moves over to the 'em,' so instead of 'git em,' it becomes 'gidem."WOMAN'S VOICE: "Gidem asoon as possible. Gidem asoon as possible." (If we move the other final consonant sounds followed by vowel sounds, and apply reduced forms and sentence rules, it would sound like this: "Gide ma soo nas possible," Nina Weinstein says.)RS: "How do you go about teaching this? What would you suggest?"NINA WEINSTEIN: "First of all, I would suggest starting with the first rule: final consonant sound followed by a vowel sound. So when people were in class and they were listening to tapes of natural speakers, I would write a sentence on the board, for instance. And then analyze it, and ask them before they even thought about it or heard it or whatever, if there's a final consonant sound in that sentence followed by a vowel sound, and how they think it might be said. And then play the sentence and see if it is said that way. If a native speaker says the sentence without stopping, the final consonant sound followed by a vowel sound is almost impossible not to do."AA: English teacher Nina Weinstein is the author of "Whaddaya Say? Guided Practice in Relaxed Speech," published by Longman. You can find transcripts and audio files of all of our segments -- including this one and our recent discussion of reduced forms -- at our Web site, voanews.com/wordmaster.

RS: And that's Wordmaster for this week. Our e-mail address is [email protected]. With Avi Arditti, I'm Rosanne Skirble.