Skip Navigation | ANU Home | Search ANU | Students | Staff

TEXT ALOUD - INEXPENSIVE AND VALUABLE

by Andrew Downie,
Information & Technology Services, OTEN-DE

This is another of those shareware software packages which could be helpful to many students. It may also limit the number of headache pills required by those charged with supplying learning material in alternative formats.

Text Aloud is a Windows application. It works with text files and text copied to the Windows Clipboard from other applications. Having opened the required file or loaded the contents of the Clipboard, clicking on the Read Aloud button causes the material to be read via synthetic speech. Microsoft Speech is the basic offering, the voice supplied being Mary. Other voices can be downloaded from the web site (see below) if you don't get on with Mary. If you want much better speech (the best we've heard so far), the AT&T speech is available for extra financial consideration. More about that speech and financial and other considerations later.

Figure 1: TextAloud Window

As each word is read, it is (optionally) highlighted on the screen. There are potential benefits here for people who have reading difficulties, although one of our testers found the feature somewhat distracting. You can adjust the size and style of the font to suit individual wants/needs. Reading can be paused and resumed at any point. By right clicking or using the Windows Application Key, options such as reading highlighted text and reading from current position are also available. Reading speed can be adjusted over a huge range.

Figure 2: TextAloud Window showing highlighting as text is read.

Having text read aloud is useful but that's not the end of the story! Another option is to have the material read to a file. This can either be a wave (standard Windows audio) file or can be in MP3 format. The latter is much smaller than a wave file, the exact size depending on the quality you choose. The file can be played back later with audio software such as Windows Media Player and the wonderful Dictation Buddy reviewed in the previous News Sheet. Especially if software capable of editing (such as Dictation Buddy) is used, there is huge potential. Material can be played, replayed, edited, sped up, annotated etc. It could be copied to a cassette, audio CD or CD-ROM. The possibilities should get you excited!

While Text Aloud is reading to a file, it makes no sound. The text scrolls through and you're given a readout of the percentage of the file processed. You can go on with other work while the reading is being done. But here's the really clever bit! Material which takes - say - an hour to listen to will take an hour to record, right? Not necessarily. You can record in "real time" if you wish. On the other hand, you can record at up to 50 times normal speed. That is, an hour's worth of listening could be recorded in (reaching for the calculator) just over a minute! The manufacturer issues a caveat to recording at very high speed in that it may cause various bits of your computer to squabble, resulting in a crash. We have tested at settings up to 50 x without crashes, but didn't achieve much more than 10 x - still not bad as a labour/time saving device. The manufacturer suggests working up gradually from 10 x while you discover the whims of your computer.

Play standard TextAloud Voice (Prepared from "Read Text to File" at 10 times Speed)

Play AT&T Voice (Mike US English) (Prepared from "Read Text to File" at 30 times Speed)

Play AT&T Voice (Audrey -UK English) **NEW** (Prepared from "Read to Text File" at 50 times Speed)

Installing the program is a breeze. By and large, so is using it. We would prefer, though, that more of the settings were available from the top menu. While some are available from there or by use of shortcut keys, quite a few require access to options panels with the mouse. For those of us who prefer keyboard controls, this detracts from the package. It can also result in some features being illusive. For example, despite the quite good in-built help, the speech rate adjustment feature remained hidden for some time (some threats of violence eventually did the trick).

Now to the AT&T speech. As mentioned earlier, this is the most impressive synthetic speech we've encountered. It is very clear and pronunciation is, by and large, quite accurate if you call an American accent accurate. There is one male and one female voice, Mike and Crystal.

There are a couple of issues to consider if contemplating use of the AT&T speech, both related to computer limitations. You need 1gb of hard drive space to load the 16khz version and they recommend 256mb of RAM. We have run it on a machine with 128mb. The result was very good speech, but the computer had to think for some time before speaking. When writing to a file, the hard drive was also very busy. Even with 282mb on board, very large files read aloud caused Mike and Chrystal to get flustered, resulting in some jumbled speech. An 8khz version is available and, judging from the sample on the website, is quite reasonable. Demands on the system should also be less.

All in all this is an extremely valuable and easy-to-use piece of software. Its potential benefit to both those who have reading difficulties and to those producing alternative format material is huge. Its price is also very reasonable, at $25US. The AT&T speech is an additional $25US. It is produced by Nextup.Com. You can get it (and other interesting-sounding software) from their website - www.nextup.com.

textAloud