About this Website

When I started learning Chinese a few years ago I looked around for a website (or a program, or a book and CD set) like this but found nothing exactly like what I was looking for. The idea is an obvious one, but until now, with improvements in speech synthesis technology, it would have been impractical for the sheer number of recordings you would have to make using human voice actors. The total number of recordings to represent all the two character pinyin combinations in these questions is 46,985. Having said that, I actually think using a speech engine is even better than a human actor for this kind of exercise because the essence of a minimal pair is that its two parts have to differ in only one phonological aspect, something a human actor could never consistently achieve.

The early version of this website used the Microsoft text to speech API that shipped with Windows, and I have now supplemented that with the AI based Azure speech API which does a better job of neutral tones. As the technology for generating the sound files gets better, and as I get better at using it, I will continue to update the recordings.

I feel that the website is now mature enough to start charging a small usage fee to cover the server costs. I have set this initially at $10 per year. Every account created before 16 October 2026 was given a free year's subscription. Your Settings page shows details.

The translations of the website from English have been done with the help of AI. Please let me know if you see anything that needs correction.

About Me

I am originally from the UK but since 2001 I have lived and worked in Spain where I am a programmer and database designer in an energy company. I have been studying Chinese since 2007. I have only ever been to China as a sightseer and this is what is behind my interest in devising tools like this for developing speaking and listening skills without prolonged immersion.

- Richard Jefferies Valencia, Spain