The trainchinese system has been specially designed for learning the Chinese language.

Choose which aspects you want to learn

Some people want to learn to read Chinese characters, while others are only interested in understanding spoken Chinese. The trainchinese system lets you focus on the aspects you're interested in. Each flash card that you create can be trained in several training modes: Reading, Audio recognition, Translation and Dictation.

You can pick default modes that are applied to all new flash cards (see your training settings) and also adjust modes for individual cards later on.

The Reading and Dictation (writing) modes are available for both Simplified and Traditional Chinese. Traditional characters are still used in Taiwan and Hong Kong, while Simplified variants are used in mainland China and Singapore.

Increased effectiveness by focusing on problem areas

The system remembers which modes you have difficulty with and presents flash cards repeatedly in those modes until they're passed. If you have trouble reading Chinese characters but none recognizing spoken words, for example, then only the reading exercise will be repeated for a flash card – resulting in a more effective learning process and a lot of time saved.

Spaced repetition

To memorize a large number of words it is important to keep working on them for a long period of time. Spaced repetition is a proven method for memorizing vocabulary that helps you organize that learning period. Here's how it works. You learn a word once, then put it aside and repeat the process several times at increasing intervals. We call these intervals "sleeping time".

Young children may learn new words after being presented with them on just a few occasions with a relatively long sleeping time between each, while adults often require many more repetitions at shorter intervals.

In the trainchinese system, you can configure both the initial sleeping time and also the number of times they'll "re-awaken" for further training. The sleeping time increases after each repetition.

Typically, people who memorize words quickly select a lower number of repetitions with a longer initial sleeping time, while others (including the older learners among us) may choose more repetitions and shorter sleeping times to boost memorization. The graph on the right illustrates this effect.

How sleeping time is calculated

There are two algorithms for calculating sleeping time as flash cards are repeated - and you can pick whichever you feels works best.

The standard algorithm

Hermann Ebbinghaus was a psychologist who described in 1885 the concept of the "forgetting curve", whereby the ability to recall information diminishes after you've learned it in an exponential pattern. Simply put, your ability to recall a word learned halves after a matter of days, and halves again after the same length of time. Before long, you can't remember it at all! The only defence is to re-learn the word regularly, at suitable intervals. In accordance with this theory, our standard algorithm doubles the sleeping time after each repetition of the word until you've learned it for good.

Using this algorithm, a fast learner who picked 7 days as their initial sleeping time and opted for 3 repetitions would train a flash card as follows:

• Learn it once on day 0
• Card sleeps until day 7,
• Re-learn it on day 8,
• Card sleeps until day 22
• Re-learn it on day 23
• Card sleeps until day 51
• Finish learning on day 52

A slower learner who picked 2 days as their initial sleeping time, with 5 repetitions:

• Learn it once on day 0
• Card sleeps until day 2,
• Re-learn it on day 3,
• Card sleeps until day 7
• Re-learn it on day 8
• Card sleeps until day 16
• Re-learn it on day 17
• Card sleeps until day 33
• Re-learn it on day 34
• Card sleeps until day 66
• Finish learning on day 67

The advanced algorithm

The advanced algorithm works in the same way, but with two fundamental differences. Firstly, it adds a training mode-specific review. Secondly, rather than always doubling the sleeping times, they are adjusted based on how well you knew the flash card when you trained it.

The review stage works like this. Let's say that you've just passed a flash card in all its active training modes, but in Reading mode you had to see the card a couple of times before you answered correctly. The card would sleep as usual, but would then wake after a short period to be reviewed in Reading mode. Once you've completed that review (passed the card in Reading mode) the card will sleep properly until it's ready to be learned again in all the modes. Additionally, the sleeping times will be continually adjusted based on your overall performance on the flash card, so that it may only sleep for 50% of the time you'd expect from the standard algorithm (doubling with each repetition), or up to 150% of that time if you always answered it correctly.

Overall the advanced algorithm increases the retention of words by adding more check points, as well as reducing the time spent learning words you already know well. but it'll certainly take longer to complete the learning of more difficult words.

As a guide, we'd recommend that most people stick with the standard algorithm, while those who have trouble memorizing the words will learn better with the advanced algorithm.