Chinese Dialogues

If you want to improve your listening and speaking in L2, you will have to do a lot of listening.

This page will explain how you can generate lots and lots of i+1 content by making your own mp3s to learn from.

You will need a native speaker and a recording device.

You can rely on content that you produce in this way because it comes from a native speaker. Additionally, it completely frees you from the shackles of lousy romanization systems – you don’t need to know how a word is romanized, or what tone a character is – only how it sounds.

I know there is a strong desire on the part of the learner to not listen to their own voice, but I strongly recommend overcoming this fear, because it’s not actually a problem[1].

The Method

You get your native to go through the list of related words, with descriptions and example sentences where necessary. For example, you could go through 車-related compounds – 車呔 , 軚盤, 擋風玻璃, 車燈 etc.

You can ask them specific questions about what you would like to learn, preferably in Chinese. You can also ask for clarification whenever you don’t get something, or tell them to slow down/speed up etc.

Make sure you record everything – you can listen to your i+1 content as many times as you like, and pick up vocab, pronunciation, grammar and so on.

Concrete Example

I recorded many dialogues about Mandarin compounds with a particular character: for example, 入口/entrance, 出口/exit, 口號/slogan, 口味/taste, 讚不絕口/praise endlessly, 口若悬河/speak skillfully.

My native and I usually fit 3 to 4 two-character compounds into a 3-minute time-slot. 成語 take up a whole three minutes, because they are harder to understand and remember.

Content is typically something like this:
我:這個…是不是入口呢?
她:對啊!你知道這個餐廰的入口在迬哪裡呢?
我:知道!在那邊! (Points at the door.)
她:沒錯!
我:你也有幾個入口耶…
她:找死是不是?我們說其他的吧! 那…你聽過「口號」嗎?
我:沒有…「口號」是什麼?
她:……

It goes on like this.

Further Comments

1) When recording, I will usually be quite passive. I will just ask ‘這個字怎麼讀?’ or ‘可以解釋這個嗎?’ The time might be split 20:80 between me and the native, so I only end up listening to myself a small percentage of the time.

2) After a while, I got used to my own voice. I think most people dislike listening to themselves, but after a while, you’ll stop being bothered by it.

3) I tried native monologues before, but I found they tended to be much less interesting to listen to. Also, because they included no interaction on my part beyond selecting learning material, I tended to understand less, and it was difficult for the tutor to gauge the difficultly level correctly.

In any case, you can buy CDs of monologues and dialogues for Cantonese and Mandarin, if you don’t want to pester a native.

4) I will always add some background music. This is a great investment of time – it might take a minute to add music to each mp3, but it will mean that you will be able to listen to it for months rather than days.

I typically pick music from Nintendo games from my childhood, which further increases the desire to listen back to L2 later, although I have also experimented with jazz, classical music and songs in languages I don’t speak.

5) These mp3s are especially effective at clearing up interference in characters with tones I’m unsure of.  For example, I kept forgetting that 漲 was pronounced zhang3 in Mandarin.  Therefore, we recorded a three-minute mp3 in which we discussed 水漲, 通漲 and 漲價, mostly in Mandarin.  After carefully listening to the recording a few times, I can now get the pronunciation correct when reading Chinese newspapers.

6) I sometimes add vocabulary learnt in this way to Anki, and sometimes not. It depends on my mood.

7) It may take quite some time to perfect your recordings. You will have to patiently explain to the native what it is that you want, and give constructive criticism as necessary.

“I think you could speak a bit slower next time”

“Perhaps you could repeat the word a couple of times slowly at the start before talking about it?”

Good Luck!

[1] Possible reasons: Your native speaker corrects you when necessary; your brain has some filter which screens out stuff which it knows might be wrong; you spend so much time focusing on what your native is saying when listening that you have no time to listen to yourself; on recordings, your native speaks for much longer than you, so statistically you hear good L2 more than not-so-good L2; you can hear that you said something wrong when listening, and you mentally correct yourself each time; you edit out serious mistakes using Audacity…