Spezialseiten > Android
I) Purpose of the app
The app is designed to give visually impaired people the opportunity to learn vocabulary orally, hence the "or" in the name. It's all done via voice input and output. Otherwise, there are just a few large buttons. Better vision is only required for the settings.
II) Current status
I originally developed this app for my mother; she wanted to learn more Turkish vocabulary. With only about 2% vision, this is difficult with conventional means. So I wanted to develop a solution that works with limited vision. Unfortunately, she is also hard of hearing, so she had trouble understanding individual words correctly. What I'm missing is a beta tester. If there are users interested in this app, I'm happy to invest more time in it. The app is and will remain free. The main problem is finding vocabulary lists. I'm happy to help with the adaptation described below, as long as I don't run into any copyright or similar issues.
There are three problems I can't solve myself: the quality of the language output and input, and especially the vocabulary lists. Most people who want to use this app have already set up their smartphone or similar device in such a way that they have partially solved the first two problems. As you can imagine, it's not possible for me to create vocabulary lists for all possible language pairs. So you have to use existing ones from the internet. Unfortunately, there is no standard format. For this program, I used Excel files with each word in a column. In addition, you have to add three lines at the top with information about the file, e.g., which language is in which column, etc. More on this below. I will try to find a partner (publisher or similar) who could provide vocabulary, ideally for free. As long as I don't have anyone, I'm of course willing to help. Of course, you have to be careful not to commit any copyright infringements. For your own use, downloading a list, adding the three lines, and importing it into the app shouldn't be a problem.
III) Concept
Suppose you're taking a German-English course. The program then creates two lists with the word pairs. Each word pair is entered multiple times (5 times by default).
You can then choose whether the German or English word should be read aloud. The other direction is then asked. If you answer correctly, the word is deleted from the list, but only once for that direction. So, for example, you have to answer it correctly 5 times before it disappears from the list completely. This repeats until you cancel it or the list is empty. It's like working with cards the classic way, placing 5 copies on the stack at the beginning and removing that one card when you get a correct answer. Since voice input isn't always reliable, or depending on the vocabulary list, the app can't always clearly determine whether the answer is correct, the user can veto a supposedly incorrect answer, and the answer is accepted as correct. In order to get by with as few and as large buttons as possible, some of them have double functions, but always two similar actions, e.g. reading out help for the current situation or displaying this written help.
IV) Usage
The following "screens" are available:
1) Main screen with 3 phases
2) Course selection
3) Settings
4) Text display
5) Help
When the program starts, the main screen normally opens during the startup phase. Unless a course has not yet been selected, then the course selection opens first.
1) Main screen
It has the same structure in all three phases: two lines of text at the top, followed by four buttons in two rows.
Except in the start phase, the text lines serve to display the vocabulary, as an alternative to reading aloud.
Since space is limited, if not everything is visible, you can click on one of the lines to go to the text display, where only the two texts are displayed.
In the start phase, the total number of words in the list and how many still need to be practiced in both directions is displayed, i.e., by default, five times as many as words at the beginning.
At the bottom left is the button with the lightbulb for help. A normal click on it provides brief verbal help for the current situation; a long click displays this text. To return to the start phase from the question or correction phase, you must press the back button.
1a) Main screen: Starting phase
The two top buttons are used to start the training. The top left button reads your native language and requires you to translate, while the top right button does the same the other way around.
The bottom right button takes you to settings, loading courses, and so on. This button has two functions. A normal click takes you to course management, a long click takes you to other settings.
1b) Main screen: Queastion phase
If you press one of the top buttons during the start phase, you'll land here. First, the word is read aloud in the source language. Then you have as much time as you like to think about it. If you want to hear it again, click on the top left. If you think you know the answer, click on the top right. Then you can say the answer. If the app thinks it's correct, it goes straight to the next word. Otherwise, or if you click on the bottom right, it goes to the correction phase.
1c) Main screen: Correction phase
Here, you have the veto option mentioned above.
First, the word is read aloud in the target language. Here, too, you can repeat it as many times as you like using the button in the top left. Then you have to decide for yourself whether you knew the answer or not. If you did, click the top right button, otherwise, click the bottom right button. In both cases, you go straight to the next word and thus back to the question phase.
2) Course selection
This is where you'll find the largest number of buttons, namely 6. The app saves the data in its own XML file format in the app's private directory. Since you can't find any files in this format online or anywhere else, you have to import the data in a different format. So far, only xls is supported. I could add other formats if needed. The button for this import is at the top left. Next to it is the export button, which is currently mainly used for troubleshooting etc. Below it is the name of the currently selected file. Below that are two buttons for switching to the next/previous file.
The help is at the bottom left, as usual in this app. Again, it's read aloud or written, depending on how long you click. At the bottom right, you can delete the currently selected file. Be careful, this is final. To return to the main screen from the course selection, you have to press the back button.
3) Settings
There are a few settings here; the standard settings dialog is used, which is not optimized for the visually impaired. Alternatively, I could have tried setting the values using voice recognition, but that's too error-prone for me. So if you really want to change something here, you should ask someone with better eyesight. First, you can specify how often a word is copied into the list at the beginning (default 5). Then you can specify what happens if you give an incorrect answer. Should the number stay the same (default = 0) or even increase? These values can be set separately for each direction. To get back to the main screen from here, you have to press the back button.
4) Text display
Because the main screen is optimized for listening and speaking, there's very little space left for the written display. If you really don't understand what's being read and can still read a bit, it's still there in writing, but often there isn't enough space. If you click on one of the texts in the main window, you'll land here. Aside from the usual help button, the entire screen is reserved for the two texts, and you can even scroll if necessary. To return to the main screen from here, you have to press the back button.
5) Help
This detailed guide is displayed here. To return to the main screen, press the Back button.
V) File formats
Here the visually impaired user probably needs help.
1) xls-File
Typically, the words of each language are on a single row. You can find many vocabulary lists like this online. However, the language you want first isn't always in column A and the other in column B. Sometimes, for example, there's a sequential number in column A... Furthermore, the app can't recognize which languages are involved or how many words there are. Therefore, you have to add three additional rows at the top. They are structured like this:
vokabelnor | 2 | 1 | 0 | 1000 | Spanisch |
| ES-ES | DE-DE | ||||
| Spanisch | Deutsch | ||||
| azul | blau | ||||
| aqui | hier |
The first line contains basic information: first vokabelnor to identify that it's really a file for this app, then the version number of the file format. This should currently always be 2. If these two fields are incorrect, the process aborts. Then come the column numbers for the two languages.
Here, language 1 is in column 1 and language 2 in column 0 (as is usual in computer science, counting starts from 0). Then come the number of words and the name of the lesson. This is a Spanish lesson with 1,000 words in columns 1 and 0.
From the next line onwards, only these two columns are used. First, a column with the language abbreviations according to the BCP47 format, see, for example, here. Specifically, the one with the language and country, e.g., DE-DE for German German or DE-AT for Austrian German. These abbreviations are very important because they are used to configure the voice input and output. If you make a mistake here, the voice output will sound strange and the input won't work. The next line is less critical. This is where the language names go in the app's language, currently only German or english. This is only for the announcement at the beginning of the training. The app uses a total of three languages: the interface, currently German and English, and perhaps French later. And the two languages of the vocabulary list. So, you can also use an English-Spanish list with the German app.
The vocabulary words start on the next line. In addition to the two texts, the number of times they will be asked is also shown.
2) VOK-File
This is the format used internally; it can currently only be written to and read from the app's private directory or exported to the download directory (then it receives the .xml extension). It is not intended to import a VOK file the other way around, as the reading routine would then require much more error-tolerant processing. So, just for the curious, here's the beginning of a VOK file:
<?xml version="1.0" encoding="UTF-8" standalone="no" ?>
<vokabelnOr
version="1"
creator="vokabelnOr - http://www.dimitri-junker.de"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">
<head> <version>1</version> <lang1>DE-DE</lang1><lang2>ES-ES</lang2><cnt>1000</cnt><langName1>Deutsch</langName1><langName2>Spanisch</langName2>
</head> <vokabeln> <vok><txt1>hier</txt1><txt2>aqui</txt2><cnt1>4</cnt1><cnt2>3</cnt2></vok><vok><txt1>blau</txt1><txt2>azul</txt2><cnt1>5</cnt1><cnt2>5</cnt2></vok>...</vokabeln></vokabelnOr>