How Phonetic Analysis Works

Shaxpir uses Phonetic Analysis to power our Rhyming Words Highlighter and our Alliteration Highlighter. These tools reveal the sounds of the words on the page, and let you see the pronunciations in bold, vivid color!

This is the Rhyming Words Highlighter in action:

And this is the Alliteration Highlighter:

But how do we know the pronunciations of all those words?

Academic Research + Open Source

The Computer Science department at Carnegie Mellon University produces an open-source Pronouncing Dictionary that provides pronunciations in a specialized syntax for more than 130,000 words, including names and places.

Here are a few examples of words from their dictionary, with pronunciations in the special syntax:

If you look closely at those pronunciations and squint your eyes just right, you can see how the whole system works. Every different sound (phoneme) in the word is represented by a distinct code, regardless of spelling (the “gh” in “laughter” creates an “F” sound), and each vowel sound creates a syllable with a numerical emphasis (syllables with a 1 are more emphasized than syllables with a 0).

You can take a look at this data online, or download the raw data to play with on your own computer, from the CMU website, here:

http://www.speech.cs.cmu.edu/cgi-bin/cmudict

Extending the Data Ourselves

Shaxpir uses the CMU dataset throughout our phonetic analysis tools, and we’re grateful to the academic community for making this open source data available.

But in spite of its extensive coverage (more than 130k words!) it wasn't complete enough to use in Shaxpir without a lot of effort to extend it. In our audits of the dataset, we found that it was missing lots of useful words (dragonflies, dreadlocks, endearments, girly, gladness, glovebox, etc, etc).

So we began a huge project to extend this dictionary with our own extensive list of additional pronunciations. To date, we have added more than 23,000 new entries to our pronunciation dictionary. This is a purely human endeavor: each of these additions to the dictionary was authored and reviewed by actual human beings.

With these additions, Shaxpir provides a truly unique and incomparable ability to visualize the sounds of speech in written text. So sign up for Shaxpir Pro today, and try out the Rhyming Words Highlighter and Alliteration Highlighter and see how it helps you in your writing!