(PGA) Use Cases (April 2013)
Steven (4th yr Undergraduate, learns better with visual, or hearing what he reads)
Steven is a forth year undergraduate student at a local university studying engineering. He has always had a knack for inventing and coming up with new ways of using things.
When he applied for university, he had initially wanted to study languages, since his family is multicultural: his father Spanish, and his mother German, both meeting on their way to Canada in the late 1960s. Both learned to speak English before he was born. He received very little exposure to his parents’ languages, raised almost exclusively with English. On occasion he would hear his parents speaking in their native tongue on calls back to relatives in Europe, and he wanted to learn to speak German and Spanish so he could understand and be able to communicate with his relatives.
He did end up taking French in high school, but German and Spanish were not offered at the school he went to. Though he enjoyed French classes he struggled learning the language. When he was speaking with his counselor about university during the his last year in high school, the councilor suggested language learning may be difficult for him, and suggested he try engineering which complimented his inventive aptitude.
In his third year of engineering, he was doing quite well in his project based work, but was continuing to have difficulty with the theory, which involves a lot of reading and written exercises. One of his professors took note of the large gap between his practical work and his study of theory and suggested he go to the school psychologist to see if she could help. His visit to the psychologist was an eye opener. While he scored quite high on visual and motor type skills, he performed very poorly on reading and memory skills, which explained a lot about why he was having difficulty with his reading assignments, and why he was scoring low on tests and quizzes. He was diagnosed with a reading disability.
The psychologist sent Steven to the study skills clinic where he met with a counselor to come up with a plan to help him get through his final year as an undergraduate and to bring his marks up so he would have a better chance of getting into graduate school. The study skills counselor suggested several cognitive exercises he could do to help with his memory, and suggested he use OCR (Optical Character Recognition) to scan his books and articles into electronic format and use a text-to-speech (TTS) program to read them aloud as he read along. The cognitive exercises focused on using visual strategies to help him reinforce what he would read in his mind’s eye, while at the same having the TTS program to enhance his own internal speech as he read. Together the two strategies made a big difference in his ability to learn from reading, improving his marks over the remainder of his third year and throughout the forth. His marks were good enough to get him into graduate school, where he would continue his study of engineering, focusing his on building a universal translator.
Now knowing better what he was good at, and not so good at, and knowing how to adapt those good skills to compensate for the not so good ones, he was confident he would make it through graduate school without any trouble, and get back to his love for languages, approaching language learning from a new perspective.
Personal Preferences that would benefit Steven
Reading:
Speak the text
Show graphics
Shorten Sentences
Spell check
Abbreviation expansion
Localization
Writing:
Dictionary
Thesaurus
Grammar check
Spell check
Memory:
Comprehension questions
Video illustration
Shorten sentences
Multimodal presentation
Summary/Table of Contents
Scenario 1: Learning German
Before he could focus on his graduate work creating a universal translator, Steven needed to learn a second language, and understand thoroughly how languages are constructed. He began taking German classes at a local German community centre not far from his university. As part of the introduction to the language, his German teacher suggested a number of tools he could use to help him learn the language. There were a number apps he could install on his smart phone, as well as several Web based services he could use.
One of the services was a Web based text-to-speech reader that included a German voice that could be used to transform the text of his lessons into spoken German. Knowing he remembers better when he hears what he reads, he installed the tool on his phone. Since much of the study materials were in electronic files, he would be able to use the TTS service to read lessons aloud while he read along.
Another app the instructor suggested was a visual dictionary that presented important concepts, and most concrete words of the language in a graphic form. Again remembering what his study skill counselor had suggested, he purchased the dictionary from the app store, and installed it on his phone.
Together the TTS service and the visual dictionary app gave him the audio and visual enhancements he needed to better comprehend the new language he was learning. While viewing text he could press on a word or sentence in his study materials, and then choose “read” to have the Web service convert the text to spoken words. Or, he could choose “view” which would search the visual dictionary for graphics that represented the words he had selected and display them dynamically over the text he was reading. Tapping the close button with graphic would hide it away again so he could continue reading.
Scenario 2: Research Reference Databases
As part of his masters thesis work Steven needed to find as much information as he could on past research associated with machine translation. Most of the databases he needed to search were available online, such as the ACM Digital Library, IEEE Xplore, INSPEC, Scholars Portal, and Google Scholar, among others.
Many of the reference databases produce a very text dense output, that for Steven was difficult to comprehend. Fortunately Steven was able to install a plugin for his browser that would dynamically reduce the output by collapsing content in sections and sub-sections, turning headings into toggles that would open or close a sections of the search results. The plugin did much the same for lists, collapsing them to a single title or label that when clicked would open or close a list. This greatly reduced the amount of information presented on the screen at one time, reducing a long list of article tiles, article information, a summary of each as well as related links for each items in the search results, to a short list of just the article titles. This made it much easier to scan and find relevant information.
When viewing the articles themselves, they were often presented as PDF documents, some with many pages of content and no table of contents at the top that would summarize the content of the article. Another plugin for his browser allowed Steven to automatically scan the PDF documents to find all the headings and generate a table of contents he could use to summarize the article. When a link in the table of contents was clicked, the browser or cursor would jump directly to a section of the article. He setup the plugin to generate the table of contents by default when any article from the research database was opened, and with the TTS software he had on his computer, he was able to scan the generated summary, and read along with the speech output to better comprehend what he was reading. He was able to record the speech output into an mp3 file, and save it to his music player so he could review the articles later and reinforce what he had earlier read.
Not all articles would work with the table of contents plugin. In many cases, where the author did not properly use headings, the table of contents would be jumbled and make little sense, or be non-existent. For some articles like this, Steven would print them off then use his OCR software to scan the article back into an electronic format. The OCR software was smart enough to know that short larger text that appeared before typical paragraph text, were headings and would generate a proper heading when the OCR’d documents were turn back into text. While this fixed the documents that had incorrect headings, it was a lot of work so he only scanned the important documents. He searched for a plugin for his phone and his computer but did not find one that would convert a PDF to text while at the same time correcting the missing or incorrectly used headings.