Technology Overview

A round-up of the most popular assistive technology, software and hardware for children and adults with dyslexia.

Introduction

Children and adults with dyslexia can now enjoy a great choice of specialist, liberating software and hardware. Technology designed for general use can also be a great help. These include word processors, electronic dictionaries and scanning software (OCR or Optical Character Recognition, which turns print into electronic text). But many specialist packages are particularly aimed at making text more accessible to those with reading and writing problems. This article gives a general outline of the types of software and hardware that are available and gives pointers to where you can find more information.

Text to Speech, making your computer talk:

Why do we want a computer to speak?

A talking computer can act as proof reader, helping you hear any mistakes or inconsistencies in your writing. It can also reduce the strain of reading: for those with concentration or reading problems hearing a long document is a lot easier than reading it conventionally. Computer speech is also important in software for teaching and is often used in spelling programs, for example to reinforce ideas using both sight and hearing.

How can you make a computer speak?

There are two ways: pre-recorded (digitised) and generated (synthesised) speech. Pre-recorded speech sounds the most natural but takes up a lot of disk space and can only read out what has already been recorded - it can’t read out something that you have just written.

Text-to-speech engines (or TTS) generate speech from text and are not limited by disk space. Both dyslexic and visually impaired people have found them to be a great leap forward in making text accessible. However, they use more or less robotic-sounding voices, which, although improving, can be grating. They are not ideal for teaching pronunciation as they will sometimes get the pronunciation wrong - words like read (reed) and read (red) can cause problems.

What text-to-speech software is available?

Text-to-speech software varies from basic screen readers where you must copy and paste text into the reader, to advanced packages which combine text to speech with OCR, to convert printed text into speech in one go. The specialist programs include other features to help with reading, spelling and word finding problems. Some of the most popular are:

Texthelp Read and Write Gold toolbar

Write:Outloud: word processor with built-in speech and phonic spell checker (available for both PC and Mac).
Texthelp Read & Write Standard: toolbars which add speech to virtually any application. Read & Write also includes a phonic spell checker, dictionary and homophone help - all with speech support.
Texthelp Read & Write Gold; Kurzweil 3000: combined text-to-speech and OCR packages to scan in documents, read them aloud and make notes, amongst many other tools.

If you want to try text-to-speech to see if it useful for you then we recommend downloading and installing BrowseAloud. Developed by Texthelp, the same company behind Read and Write, BrowseAloud reads aloud certified websites including ours and popular sites such as www.google.com and www.bbc.co.uk.

Phonetic Spell Checkers

Most word processing programs now come with integrated spell checkers and most common applications, such as Micrsoft Word, come with interactive error alert such as highlighting. But most spell checkers are looking for typing errors not spelling mistakes. Roger Mitton (1996) in his article for the Simplified Spelling Society about computer spell checkers pointed out that research has found that 80-95% of errors could be explained by simple typing errors such as omitting or inserting one letter. However this is not the case with poor spellers or dyslexic people. Many dyslexic children and adults spell phonetically, that is they attempt to spell the word as it sounds. But because the English language is not phonetically regular this does not always give correctly spelt words. Here are some examples of phonetic errors:

wose for was

fissics for physics

nowlige for knowledge

Spell checkers have been developed specifically to cope with these types of errors and are called “phonetic spell checkers”. Many come with additional features to help users identify the correct word from the suggestion list, such as speech feedback and dictionary definitions.

Phonetic spell checkers come in two types - handheld, portable spell checkers and software applications. Of all the different makes of electronic, handheld spell checkers we’ve found that only the Franklins can really cope with phonetic errors. The DMQ-570 is a pocket sized spell checker with dictionary and thesaurus, while the Franklin Literacy Word Bank, with the Oxford Primary Dictionary and Thesaurus, is designed for children in Key Stage 1 and 2. For those who need speech feedback there is the Franklin DMQ-1870 UK Speaking Dictionary that also comes with a calculator and address book.

Franklin Speaking UK Spell Checker and Dictionary, DMQ-1870

Software applications with phonetic spell checker come in two types. There are those that are built into word processors, such as Write:Outloud which uses the same spell checking routine as the Franklin handheld. And there are those that interactively check your spelling in a number of applications such as Read & Write Standard and Read & Write Gold, both of which also offer speech feedback and word prediction within word processing applications. We have a more in-depth article comparing spell-checkers along with details of the most popular Franklin Handheld spell checkers.

Homophones or Confusables

In his analysis of spelling errors Mitton also found that up to a third of spelling errors were actually real word errors, either because the word is a homophone (word that sound the same but mean different things such as see and sea) or because the error has made another word (for example from and form; then and the). These words are really difficult for dyslexic writers to pick out when proof reading. There are two methods that can be used to pick them up.

A text-to-speech application can help you identify when the wrong word has been typed by having it read back your work, but this won’t help with homophones. Read & Write Standard, Read & Write Gold and Write:Outloud offer tools to help with both problems. You get speech feedback and a tool for highlighting and correcting homophones within your document, solving both problems.

Many of the Franklin handheld dictionaries also come with a “confusable” function which gives you alternative spelling for words that sound similar but mean different things.

Speech Recognition, getting your computer to write!

Why do we want a computer to understand our speech? If a computer can understand what you are saying, it can do the hard work of turning your ideas into print. A computer has no problems with spelling and has perfect presentation - no messy handwriting ever again! Software that recognises what you are saying and transcribes it is known as Speech Recognition, or sometimes Voice Recognition. It is particularly useful for people who are better at speaking than writing and often includes basic text-to-speech for proof reading your dictation.

Speech Recognition has progressed enormously over the past decade. But it still has many limitations. Computers cannot automatically understand a voice nor can they discriminate between different voices. So to use the software you must go through a training process that involves reading a script. In the past this has meant reading for up to an hour but the new generation of software has reduced this to about 10 minutes for a normally fluent reader. Users with speech impediments have had varying success and may still prefer the older style packages where you have to speak each word separately, or discretely, i.e. with a noticeable gap between words.

Remember, too, that dictation is not the same as talking in a conversation. The computer can’t understand “umms” and “aahs” and doesn’t know much about punctuation. The latest packages can estimate full stops and commas but all other punctuation must be dictated. Dictation is a skill that has to be developed.

Speech Recognition uses a lot of processing power, so needs a modern, fast computer with a lot of memory. There are also other hardware considerations, such as sound quality, so it is vital you get advice from a specialist supplier to minimise the chance of frustration. However with support, training and time practically anyone can work with Speech Recognition, gaining the literacy skills they could never have achieved before.

What dictation software is available?

There are two main publishers of speech recognition software: Dragon and IBM. They share similar origins and you can get similar high recognition rates with either. However, overall, at the time of writing (August 2006) we feel that Dragon is the easiest to learn to use and has a far more forgiving training process.

The most popular program for educational use is Dragon NaturallySpeaking Preferred which includes text-to-speech and has a wide variety of training scripts including those written specifically for children. MacSpeech Dictate is available for the Macintosh but have much less functionality than the PC version.

For more details on which speech recognition package is suitable for you have a look at our comparison article.

Concept & Mind Mapping Software:

Dyslexic people often prefer to process information visually. So they often like to use concept mapping or mind mapping:

to organise existing knowledge and ideas;
to plan new writing and projects.

Several packages let you draw maps quickly and easily and the computer has the huge advantage over paper that it is very easy to change or develop a map without having to start again. Other advantages of concept mapping software include the ability to

change the layout of the diagram as you go along;
use images within the diagram;
use brainstorming mode to quickly build up a diagram;
spell check diagram;
export diagram or a text version of it to other programs.

Personal preferences vary, but here we’ve tried to summarise the main features of the best programs:

Inspiration (11+) / Kidspiration (5-11): flexible concept mapping programs, designed for education, which allows you to switch between a linear-text view and a diagram view. Image banks enable images to be easily added to maps. Kidspiration is a junior version of Inspiration, aimed at the primary sector. Inspiration and Kidspiration have the added advantage of built-in speech while Inspiration also has a word tool for helping with vocabulary.

MindGenius (11+): an adaptable Mind Mapping program with an automatic brainstorming mode which enables the map to be viewed in a variety of layouts. Closely integrated to Office it allows text outlines and images of maps to be easily imported and exported to many common applications. It also has time and project management tools so is particularly useful for those who need support with organisation.

Mind Manager (11+):a project management tool that allows you to draw Mind Maps and link to MS Outlook, PowerPoint and Project.

For more details visit our introduction to concept mapping.