szigetvari’s etcetera

The best things in life are free…

english pronunciation dictionary

A database of English words and their RP transcriptions was circulating in the early 1990s. I have corrected some errors and wrote a web search interface for it. After more than a decade of ignorance, I learnt from Dariusz Laska that the source is the Oxford Advanced Learner’s Dictionary 3rd ed.

I have found the dictionary quite reliable though there were (and probably still are) typos in it. If your find and send them, I’ll correct them in the database.

This dictionary can be used to look up the pronunciation of individual words and also to find spelling regularities, phonotactic constraints, and the answer to questions like “how many English words begin with ð.”

In 2011 I wrote a new interface, so that you can now filter searches by grammatical labels and frequency information too.

Geoff Lindsey and me are working on a new version that dispenses with the Gimsonian system and gives current British English pronunciations, the cube. It is frequently updated, currently giving several types of transcriptions with customizable emerging features of British English (like l-vocalization, th-fronting, etc).


In this section I share scripts I believe might be useful for others.

I wrote for displaying the time left during examinations. It’s a perl script which can show several countdown clocks at the same time, distinguished by labels and colours.

Without an argument it displays the current time in hh:mm:ss format. The first argument may be in hh:mm format, countdown will end then. It may also be in +mm format, countdown will run for mm more minutes. The optional second argument is a label for the first countdown clock. A second countdown clock may be started with the third argument, the fourth being its label, etc. (To start two countdown clocks without labels use something like 14:00 "" +70) Help is displayed if the first argument is -h, --help or cannot be interpreted as a valid time specification.

When countdown finishes, the clock simply stops at 00:00:00. Control-c quits.


These are two scripts that check whether a text file contains any other characters than ASCII 32–126. It is useful for checking TeX or HTML files after converting from some other format.

The perl script checks each input file for characters outside the standard ASCII range, giving the line and column number they appear in. The shell script nonasciitypes invokes and gives a summary of the types of nonASCII characters in each file., nonasciitypes

software and web pages I found very useful

I have found a lot of hints on people’s web pages about great software. I hope you’ll also find something useful here you didn’t know about.

I’m grateful to George Soros.

© szigetvari-AT-elte-DOT-hu