ictsan

The usual model of use for prediction is for users to have a personal lexicon that grows as they use it. The software bases its predictions on the start of what the user has typed so far in the word, but also on the words the user has already used, the number of times the words have been used, and how recently used.

Supporting the needs of such a user can be rather time consuming. For pupils who may not be long time users of prediction, it may be better to create a class lexicon or topic lexicons. It is important, using Penfriend, which usually saves a lexicon at the end of a session, that the teacher turns off this option in the ‘advanced options’ menu.

The lexicon is best  place for this lexicon is to put it in a shared area of the network where pupils have read but not write access to the lexicon. The teacher can add words in a variety of ways while being sure that pupils cannot accidentally get rid of the lexicon.

It has been clear for a while that mp3 players would be excellent devices for the visually impaired. However, many have very small displays that make them unusable by VI users. This new device is a lot more than just an mp3 player. However it is most notable for being on of the few devices of its kind with voice output to allow navigation. It has no screen and all the commands are entered via key presses. The results are heard as spoken prompts.

This is a a very impressive solution in the school setting, to the problem of delivering learning materials in audio format to pupils in a portable format. Its size is just larger than a large mp3 player or mobile phone.

It can read out text or play files in a variety of formats;

• Talking Books - These are DAISY books, which usually have audio recording of text.
• Other Books - These are mp3 files. a series of files in a folder are treated as a book of sequential sections or chapters. This would be the folder to put a text book that had been turned into audio using a voice such as Heather - see previous post.
• Textfiles - plain .txt files and webpages can be stored here, to be read out by the internal voice of the Stream
• Music - Your music mp3s go here
• Notes - You can record voice notes, or lectures using the internal mike or an external one. They are stored here.

The Stream uses standard SD cards as a storage medium. These are cheap and the contents can easily be backed up onto a spare SD card.

Generally the means of navigation has been well thought out. There are faults. The folder structure of the text files “Bookshelf” is not reflected in the navigation system. This has been pointed out to the developers and it is believed that an improvement will be implemented soon. Fortunately, upgrades are made available at no cost from the developer’s website.

The internal voice which reads the static commands is excellent. However the voice that read the filenames and the text files themselves is not as clear and requires greater concentration. A UK voice is available from the website, but it is not recommended that users upgrade to this unless a new clearer version becomes available

One last gripe; It is hard to get to grips with a whole new operating system on a device with no screen. Sighted people are not used to it. the documentation does not really live up to the ideal of providing “What you need to know in the order you need it”

At last a really good implementation of MP3 player technology well designed for VI users. Also specialist technology for VI users that is not very expensive. Early adopters are not penalised as upgrades are free and simple to effect.

A word of advice if you are supporting a young user of the Steam. Take it home for a week and use it as your Mp3 player, means of accessing podcasts and listening to blogs. It works!

.

Thanks to the CALL Centre, Edinburgh University, I have been able to use the newly licensed Scottish voice Heather. I have used it to create audio versions of text files. These have been prepared for a VI pupil who will listening to them using a Victor Reader Stream which will be the subject of a blog very soon, I hope.

Heather really is one of the clearest synthesised voices I have heard on a computer. She is a SAPI 5 voice and I have used the voice very well using Penfriend V 3.1 and Wordtalk, the free text to speech utility for MS Word. I also used a trial version of Wordtalk (v4.0) which turns marked text into a .wav or .mp3 audio file.

Here is a file of a short text that I made using WordTalk (4.0) I think sounds as good or better than any text to speech I have heard. - Wax in a candle. I improved the clarity by putting commas into longer sentences.

The text is taken from here http://wwr.wikispaces.com/Materials

Of course, the great thing is that the voice is definitely Scottish. She can manage Scottish place names rather well. Send me some text for Heather to roll her tongue round. Paste it into a comment!