Application interface.

OptimTalk Desktop Suite Professional consists of three applications – OptimTalk VoiceXML interpreter, OptimTalk CCXML interpreter and OptimTalk grammar tester – in a single, easy-to-use graphical user interface. The interface consists of a menu bar, a tab bar where the user can switch between the three applications and the rest of the main window where application frames are located.

The application frames can be re-arranged according to user's needs. All frames can be moved around the main application window, pined to a border, hidden, or moved outside the main application window. There is a menu item for resetting the layout to default if required.

In the screenshot above, the main application window is tiled between Phone simulator and CCXML interpreter frames. A dialog frame named ccxmlDialog__internal_37 is pinned to the bottom border of the application window and the Settings frame is pinned to the right border. The Logs frame is floating outside the main application window. When the mouse is moved over a frame pinned to a border, the frame slides out and hides again when the user clicks to another frame.

VoiceXML document interpretation.

At the top, the VoiceXML interpreter frame contains an input line labeled "Select VoiceXML document". The location of a VoiceXML document to interpret can be typed here. Local files as well as HTTP(S) URLs are supported. Interpretation is started and stopped by the button under the input line.

The user can listen to the audio and spoken output generated by the VoiceXML interpreter and can see all outputs in text form on the screen. At each dialog turn, the user can choose between using spoken input processed by a speech recognizer and bypassing the speech recognizer by entering the input in text form using keyboard. The text input is entered to the input line in the bottom part of the frame. DTMF input can be entered using the telephone keypad on the right side. All typed as well as recognized utterances are displayed in text form on the screen so that the user can see what exactly has been recognized as well as the complete history of the dialog.

TTS and/or ASR can be switched off if needed and the interaction can run in text form only.

Application logs.

All log entries produced during the interpretation are concentrated to a frame where they can be instantly inspected. They are also automatically stored to a file for later inspection.

The log frame consists of three parts. The upper part is the log filter where criteria for filtering the log entries can be entered. They can be filtered according to the value of an attribute or according to a combination of values of several attributes. A quick filter working immediately "as you type" is available for simple filtering.

The central part of the log frame shows a short summary of the log entries while the bottom part shows the full text of selected log entries.

VoiceXML interpreter settings.

The settings are divided into three groups. The first group contains VoiceXML interpreter related settings. In particular, the speech recognition and text-to-speech engines can be set here that will be used during the interpretation. The second group contains audio settings and the last group covers logging related options.

CCXML interpreter.

The CCXML interpreter frame provides an interface for managing active CCXML sessions. Using the buttons at the top of the frame the CCXML session manager can be started and shut down. Once the session manager is started, it waits for incoming calls and launches CCXML sessions handling the calls. A new CCXML session can be also started manually. In this case, the user must specify the CCXML document that will be interpreted within the session. Local files as well as HTTP(S) URLs are supported. The last button serves for killing an existing CCXML session.

The frame also provides information about currently active CCXML sessions. When the user clicks on one of the active sessions in the list, more information about the session is provided together with the list of connections belonging to the session and VoiceXML sessions launched from within the CCXML session. Clicking on a VoiceXML session causes opening a VoiceXML dialog frame where the user can interact with the VoiceXML dialog. (The frame is in the bottom right corner on the screenshot above.) There can be several VoiceXML interpreter frames open in the same time.

The phone simulator is used to make incoming calls to the CCXML session manager and to accept/reject outgoing calls. The phone simulator frame contains a separate phone for each call. The user can switch between them using the tabs at the bottom of the frame.

CCXML interpreter settings.

The settings contain a group of CCXML interpreter related options. In particular, telephony interface that will be used for interpretation can be selected here. Typically, phone simulator is used.

Besides the CCXML related options, the CCXML interpreter settings frame contains everything that the VoiceXML settings frame does, i.e. VoiceXML interpreter related options, audio options and logging options. The VoiceXML related options apply to a VoiceXML dialog started by a CCXML script.

Grammar tester.

The grammar tester frame contains an input line to which the location of a grammar file can be typed. Local files as well as HTTP(S) URLs are supported. Grammar root rule and grammar content-type can be specified if required.

Once a grammar is loaded, the user can enter utterances which are parsed using the grammar. If the utterance matches the grammar, the semantic interpretation of the utterance is displayed. Otherwise, an information that the utterance does not match the grammar is displayed.

Advanced settings.

The settings frame also offers an advanced settings tab where the user can enter additional settings. Each item has its name, type and value. The advanced settings are intended for experienced users who need to fine tune the VoiceXML or CCXML interpreter behavior. Using advanced settings it is possible to specify values for parameters that cannot be set using the GUI settings.

Localization.

The application can be localized to various languages. Currently two languages are supported – English and Czech.

VoiceXML document interpretation using the command line VoiceXML interpreter.

Besides the graphical user interface, the OptimTalk VoiceXML interpreter, OptimTalk CCXML interpreter and OptimTalk grammar tester are available as applications with a command line interface. This is useful e.g. for running automated tests.

The command line versions try to mimic the features known from the GUI version as much as possible. In the case of the VoiceXML interpreter, audible output is produced together with a text output on the screen. The user can choose between using speech for input and typing in the utterance using keyboard at each dialog turn. All recognized utterances are displayed in text form on the screen (lines starting with >>>). Spoken output and input can be switched off in the configuration file if needed. Telephone keyboard simulator mapped to F* keys is available for entering DTMF.

Note that these advanced features may not be available in the Basic Edition.