Setting up Audio
In the current version of Cerebware, there is no facility for recording your own audio, and it is not a high priority - most computers offer the ability to record audio anyway, and computer analysis of a learner's speech output is unreliable, at best. For speech practice, we recommend chatting to native speakers, face-to-face or online, or trying to copy an audiobook, listening critically to your own pronunciation.
The Trainer does offer audio output facilities, however, and these are highly recommended. Audio files can be in the standard Java format (with the suffix '.au'), or the well known MP3 format (with the suffix '.mp3'). Freeware is readily avalable for converting existing audio to the '.au' or '.mp3' formats, and many of the free products are also able to split audiobooks into short, sentence-length segments.
Two different audio files can be associated with each item. These files should be placed in the audio subfolder of the Vocab Trainer folder. (When referring to these files in the Edit Window, or in a list file, the path does not need to be specified). Both files can be specified in the advanced version of the Edit Dialog, as shown below:
Contextual Audio
The first type of audio is a contextual example, such as the word used in a sentence from an audiobook. The word in the example does not have to match the vocab item exactly: it may be in its declined form, for instance. (Separable German verbs may even be in their separated form, requiring the student to reassemble the primary verb and convert it into the infinitive). This sort of audio has much better mnemonic value than a basic pronunciation guide, and is highly recommended. It also provides a much less direct clue to the answer, so it can be played under circumstances when a straight pronunciation would have given the game away. In fact, during the intermediate stages of learning a word, hearing the word embedded in a sentence does not automatically lead to recogniton of the answer, and closely mimicks the natural method of language acquisition: the brain's language centres are forced to extract meaning, and use that meaning to impose word-boundaries onto the audio-input.
The best source of contextual audio is, in most cases, an audiobook, but the internet is full of downloadable mp3 files using a range of languages in a natural context, and these are very suitable. A couple of paragraphs of German news, for instance, can be used to create a themed word list, complete with audio. The files simply need to be converted to '.au' or '.mp3' format, chopped into sentence-sized chunks and linked to the vocab list, but this is not difficult. Contact Cerebware at any time for help with this process, free of charge. (For German learners, the first chapter of 'Harry Potter und der Stein der Weisen' is a good a place to start, and Cerebware has already done all the work. Any Cerebware user who obtains a legitimate copy of the audiobook should contact Cerebware to obtain a fully-processed version of the first two chapters, with sentence-long audio files.)
Pronunciation Audio
The second type of audio is the standard pronunciation of the item. This type of audiofile should be played when the item is first met, and can be replayed when testing in the Foreign-to-English direction (but it is not appropriate when testing in the English-to-Foreign direction, because it provides the answer).
Specifying Audio in a List File
The recommended way of adding audio, for beginners, is via the Edit Dialog. For teachers, however, or students wishing to share their work, it may be more convenient to add audio tags directly to a human-readable list. Within a list ('.lst') file, the two different audio files can be specified in within curly braces. For example:
ein : a {HPSW1.1.au | | | ein.au }
The first audiofile in the example, HPSW1.1.au, contains a sentence from an audiobook. The second audiofile contains a native speaker saying the German word 'ein'. Both file types are optional but, if present, the Trainer will use the two different files in different circumstances.
The three vertical lines break up the text within curly braces into four fields. Only the first and last relate to audio. The overall syntax is as follows...
foreign phrase : English {contextual audio | sentence tag | footnote | pronunciation}
Missing fields should be left blank so, to specify two audio files, enter text for the first and fourth fields, leaving an empty space for the second and third fields. If you only want contextual audio, leave out the later fields. If you only want pronunciation, provide three blank fields:
foreign phrase : English {contextual audio | | | pronunciation}
foreign phrase : English {contextual audio}
foreign phrase : English { | | | pronunciation }
(Note: If the syntax seems hard to remember, just type '{?}' within your list file and the curly braces will expand to remind you of the different fields and their positions.)
The Sound Menu
To turn audio on or off, choose an option under the 'Sound' menu. 'Automatic' sound means that the audio will play if the Cerebware program thinks it is appropriate, based on your knowledge of the item. 'Manual' sound means that the audio will not play unless you press the 'Play' button. 'New Sentences Only' means that the audio will play automatically when a new sentence is shown, but otherwise will not play unless you press the 'Play' button. Note that the Edit Dialog also allows you to play the files associated with an item.