property element
Last Updated: 09-27-2016
VoiceXML 2.0
Specifies speech parameters for an entire application, document, or form. Although the default value for each property is provided below, these values are subject to change. If your voice application relies on a specific value, set the property explicitly in your application root document.

Syntax

<property
expr = "ECMAScript_Expression"
name = "string"
value = "string"
/>

Attributes
expr
An ECMAScript expression that evaluates to the value assigned to the property. This attribute is a Tellme extension.
name
Required. The name of a property to set.
audiofetchhint Specifies whether or not audio files may be pre-fetched. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
prefetch Audio files may be prefetched.
safe Audio files may only be fetched when needed, never before.
audiomaxage The maximum acceptable age, in seconds, of cached audio resources. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
audiomaxstale The maximum acceptable staleness, in seconds, of expired cached audio resources. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
bargein Controls whether or not a user can interrupt the system. The property applies to both DTMF and spoken input. For more information on controlling bargein, see Understanding bargein.
false Disables the user's ability to interrupt queued prompts by speaking during audio playback. Disabling barge-in may displease users, but it can used effectively when you want to ensure the user hears a prompt in its entirety.
true Enables the user's ability to interrupt queued audio during audio playback.
bargeintype When bargein is enabled, this property controls the type of bargein that will be performed in response to voice or DTMF input.
hotword A prompt will continue playing until either a complete match of an active grammar is detected or there is no input during the time specified for the interpreter to listen for input.
speech A prompt is stopped as soon as speech or DTMF input is detected, irrespective of whether the input matches a grammar.
completetimeout A time designation specifying the length of silence that is required before the speech recognizer finalizes a result (either accepting it or throwing a nomatch event), when the speech is a complete match of an active grammar. The valid range for this property is 0.1 seconds to 10.0 seconds. The completetimeout value should be less than the incompletetimeout value otherwise unexpected behavior may occur. If the completetimeout value is out of range, or both the completetimeout and the deprecated tellme.endsecondsonuniquematch properties are specified, the interpreter throws a generic error event. The default value is 0.25 seconds. This property has no effect prior to Revision 3.
confidencelevel A confidence level from 0.0 to 1.0 inclusive. If the confidence score for an utterance is below the specified level, the utterance is rejected. The value must be specified as a floating point number (e.g., use 1.0 to specify the maximum). The default is 0.5.
documentfetchhint Specifies whether or not documents may be pre-fetched. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
prefetch Documents may be prefetched.
safe Documents may only be fetched when needed, never before.
documentmaxage The maximum acceptable age, in seconds, of cached documents. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
documentmaxstale The maximum acceptable staleness, in seconds, of expired cached documents. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
fetchtimeout A time designation representing the interval to wait for a resource to be fetched before throwing an error.badfetch event. The fetchtimeout value is rounded down to the nearest whole second (e.g., 5700ms or 5.7s would be rounded down to 5s). If the fetchtimeout property has a value less than 1 second, it is reset to the default value. The default value is 15 seconds. If the value specified is invalid, the interpreter throws error.badfetch. This property has no effect prior to Revision 3.
grammarfetchhint Specifies whether or not grammars may be pre-fetched. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
prefetch Grammars may be prefetched.
safe Grammars may only be fetched when needed, never before.
grammarmaxage The maximum acceptable age, in seconds, of cached grammars. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
grammarmaxstale The maximum acceptable staleness, in seconds, of expired cached grammars. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
incompletetimeout A time designation specifying the length of silence that is required to decide the end of speech when the last recognized word in the utterance is an incomplete match to an active grammar. The valid range for this property is 0.1 seconds to 10.0 seconds. The completetimeout value should be less than the incompletetimeout value otherwise unexpected behavior may occur. If the incompletetimeout value is out of range, or both the incompletetimeout and the deprecated tellme.endsecondsonallowablematch properties are specified, the interpreter throws a generic error event. The default value is 1 second. This property has no effect prior to Revision 3.
inputmodes Controls the type of input the speech system should accept. Use this property in conjunction with the mode attribute of the grammar element to optimize application performance.
dtmf The speech system should accept DTMF input only.
dtmf voice The speech system should accept both DTMF and spoken input.
voice The speech system should accept spoken input only.
interdigittimeout The amount of time the Platform waits between each DTMF input character before interpreting the DTMF string. The default value is 3000 milliseconds. Prior to Revision 3, the interdigittimeout value is assumed to specify the number of milliseconds, but in Revision 3 and later it is a time designation.
maxnbest Controls n-best processing by setting the maximum number of items in the application-scoped lastresult$ array defined after a successful recognition. The default is 1 which means that n-best processing is disabled.
Be judicious when setting the value of this property since large values negatively impact recognition performance. Every time the value of maxnbest is doubled, recognition performance decreases by 4%. For example, setting maxnbest to 128 results in roughly 22% slower recognition compared with setting the property to 4.
maxspeechtimeout A time designation representing the maximum duration of user speech. If this amount of time elapses before the user stops speaking, the maxspeechtimeout event is thrown. The default value is 60 seconds. This property has no effect prior to Revision 3.
objectfetchhint Specifies whether or not objects may be pre-fetched. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
prefetch Objects may be prefetched.
safe Objects may only be fetched when needed, never before.
objectmaxage The maximum acceptable age, in seconds, of cached objects. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
objectmaxstale The maximum acceptable staleness, in seconds, of expired cached objects. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
recordutterance Causes the VoiceXML interpreter to record the user's utterance and to set three shadow variables on the appropriate form item variable: recording, recordingsize, and recordingduration. The interpreter also sets the corresponding properties on the application.lastresult$ object. For more information, see the utterance recording tutorial.
false Disables utterance recording.
true Enables utterance recording.
recordutterancetype The format to be used for utterance recordings, if utterance recordings are enabled via the recordutterance property.
audio/basic Raw (headerless) 8kHz 8-bit mono mu-law [PCM] single channel.
audio/x-alaw-basic Raw (headerless) 8kHz 8-bit mono A-law [PCM] single channel.
audio/x-wav WAV (RIFF header) 8kHz 8-bit mono single channel.
scriptfetchhint Specifies whether or not scripts may be pre-fetched. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
prefetch Scripts may be prefetched.
safe Scripts may only be fetched when needed, never before.
scriptmaxage The maximum acceptable age, in seconds, of cached scripts. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
scriptmaxstale The maximum acceptable staleness, in seconds, of expired cached scripts. There is no default value. If the value specified is invalid, the interpreter throws error.semantic. This property has no effect prior to Revision 3.
sensitivity The sensitivity to noise during input recognition. This affects whether bargein might be triggered by noise rather than actual user speech. The valid range for this property is 0.0 (least sensitive to noise) to 1.0 (highly sensitive to quiet input). Thus, if you set the property to a low value, the recognizer is less sensitive to noise, but the user must speak more loudly in order to be recognized. If the sensitivity value is out of range, or both the sensitivity and the deprecated tellme.bargeinlevel properties are specified, the interpreter throws a generic error event. The default value is 0.5. This property has no effect prior to Revision 3.
speedvsaccuracy The desired balance between speed and accuracy. The valid range for this property is 0.0 (fastest recognition) to 1.0 (best accuracy). If the speedvsaccuracy value is out of range, or both the speedvsaccuracy and the deprecated tellme.pruning properties are specified, the interpreter throws a generic error event. The default value is 0.5. This property has no effect prior to Revision 3.
tellme.bargeinlevel DEPRECATED. In Revision 3 or later use sensitivity instead. tellme.bargeinlevel specifies the volume level in decibels (dB) from 5 to 30 at which a user needs to speak in order to trigger bargein. If you set the property to a large value, the recognizer is less sensitive to noise. Consequently, however, the user must speak more loudly in order to be recognized. The default is 12.
tellme.confidential
false Utterances are logged if utterance logging is enabled.
true Utterance logs will not contain the recognition results and recorded utterances will not be persisted. Grammar URLs are always logged.
tellme.dtmfbufferexpiretime A time designation representing the number of seconds the VoiceXML interpreter looks back for the last digit pressed from the current state to determine whether to discard the entire DTMF input buffer. The property is only applicable if bargein is enabled. Buffering of DTMF input can be useful to allow experienced users to speed through a sequence of dialogs. While there is no upper limit on the value, you should keep it reasonably small to prevent queueing of extraneous user input. The default is 4 seconds.
tellme.endseconds A floating point value indicating the number of seconds of silence the end-pointer must detect before signaling end-of-speech to the recognizer. The default value is 1.0. If the value is too small, the system may cut off callers before they are finished speaking, typically resulting in a nomatch event. If the value is too large, the system will seem slow to the user due to gaps of silence between when the user finishes speaking and when the VoiceXML interpreter plays the next prompt. For classes of utterances that result in natural pauses, such as when stating a phone number, set tellme.endseconds to 1.0. If the recognizer is able to clearly accept or reject the user's utterances, it can override the end-pointer even though the end-pointer has not explicitly signaled end-of-speech. The tellme.endseconds property only impacts timeout for speech. Use termtimeout to customize the behavior of DTMF input recognition.
tellme.endsecondsonallowablematch DEPRECATED. In Revision 3 or later, use incompletetimeout instead. The number of seconds of silence required to decide the end of speech when the last recognized word in the utterance is an allowed final word in the grammar. The valid range for this property is 0.1 to 10.0. The default value is 1.0.
tellme.endsecondsonuniquematch DEPRECATED. In Revision 3 or later, use completetimeout instead. The number of seconds of silence required to decide the end of speech when the last recognized word can only be a final word in the utterance in order to match the grammar. The valid range for this property is 0.1 to 10.0. The default value is 0.25.
tellme.field.recordutterance DEPRECATED. Use recordutterance instead. Causes the VoiceXML interpreter to record the user's utterance and to store a reference to it in the recording field shadow variable. The utterance is recorded regardless of whether the recognition succeeded or resulted in a nomatch.
false Disables utterance recording.
true Enables utterance recording.
tellme.incrementalgarbageweight DEPRECATED.Used to optimize the rejection of out of grammar utterances, also known as the garbage model. The weighting specified by this property is applied once for every 10 millisecond frame. The valid range for this property is -200 to 0. The default value is -50. Smaller negative values make it more likely to match the garbage model (@reject@) grammar.
tellme.magicword DEPRECATED. In Revision 3 or later you must use bargeintype instead. Controls the system's ability to continue playing a prompt while simultaneously comparing a user's utterance to an active grammar. Magic word recognition is extremely robust against noise but causes the system to respond more slowly. Only enable the property for lengthy content delivery, such as a weather or a news report. If a user speaks or types DTMF keys during the delivery of content, the system continues to play it while the recognizer compares the utterance to the active grammars. The complexity of the active grammar(s) has a direct impact on the response time, so keep the grammars as small as possible. DTMF input overrides the magic word property unless the inputmodes property is set to voice. By default, if the DTMF input doesn't match the active grammars, the Platform throws a nomatch event. tellme.field.recordutterance and tellme.magicword are incompatible. That is, if you enable utterance recordings, the interpreter automatically disables 'hotword' bargein.
false Disables magic. If the bargein property is enabled, and the system is waiting for user input, an utterance from the user is matched against the active grammar(s), resulting in a recognition or a nomatch event.
true Enables magic causing barge-in to operate silently. The start of speech does not interrupt the current prompt, and if an utterance is not recognized, the system continues playing the prompt rather than responding with a nomatch. When a phrase is recognized, however, the VoiceXML interpreter stops playing the current prompt, and executes the VoiceXML corresponding to the utterance.
tellme.phoneticpruning Controls the pruning of unlikely hypotheses during recognition by performing additional computation. When using this feature for large grammars, recognition can take significantly less time while sacrificing some accuracy. For small grammars, however, your application's recognition and accuracy performance can be negatively impacted. Enable this feature only when dealing with large grammars, and you determine via thorough testing that recognition performance improves significantly.
false Disables phonetic pruning.
true Enables phonetic pruning.
tellme.pruning DEPRECATED. In Revision 3 or later, use speedvsaccuracy instead. The width from 400 to 5000 of the Viterbi search over the acoustic models. The search includes the number of paths to consider for each iteration. Use this parameter to fine-tune the trade-off between accuracy and speed. A value of 5000 provides the most accurate but slowest response. A value of 400 provides the fastest but least accurate response. The default value is 1200.
tellme.timeoutondtmf DEPRECATED. Use termtimeout instead.
tellme.wordtransitionweight Used to optimize the acceptance of each matched word. The valid range for this property is -1000 to 0. The default value is -200. Large negative values penalize phrases with many short words, and favor longer words. A zero value indicates no word insertion penalty.
termchar The terminating DTMF character when recognizing DTMF input. The default value is "#". To disable the terminating DTMF character, set the value of this property to a single space character (" ") in revisions prior to revision 3, or to the empty string ("") in Revision 3 and later.
termtimeout A time designation representing the terminating timeout value when recognizing DTMF input. As the user enters DTMF input, the interpreter matches the accumulated input against the active grammars. Ultimately, the caller may have typed enough input for the interpreter to match a complete grammar entry. Once this happens, the interpreter activates the termtimeout timer. If the value is 0, the default, the interpreter returns the return value from the grammar immediately. Otherwise it accepts additional input that may match against a different grammar entry or result in a nomatch event.
timeout A time designation representing the interval of silence following the end of the last prompt that the interpreter allows while waiting for user input before throwing a noinput event. The default is 7 seconds.
universals Controls default Platform command grammars. Default Platform event handlers are not affected by this property.
all Enables Platform default.
none Disables Platform defaults.
value
The value to assign to the property. Values are specific to the property.

Parents
field, form, initial, menu, object, record, subdialog, transfer, vxml

Children
None.

Remarks
The property element lets you specify default speech parameters for an entire application, document , form , or field .
Properties set within the innermost container take precedence over properties set within outer containers. For example, properties set within a field take precedence over properties set within the form, document, and application.
The expr and value attributes are mutually exclusive.
Properties are evaluated each time the VoiceXML interpreter enters a listen state. Use the expr attribute sparingly since numerous executions of the ECMAScript interpreter upon entry into a listen state will have an impact on application performance.
Properties that accept time designations should be set to a non-negative real number followed by a time unit identifier - either "ms" for milliseconds or "s" for seconds. For example, "3s" represents 3 seconds; "3000ms" represents 3000 milliseconds, and "1.5s" represents 1.5 seconds.
Examples

The following example reads back a fictitious financial news story to the user. The bargeintype property is set to "hotword" because the story is lengthy, and we only want to interrupt the story if the user says "stop." The timeout property is set to zero because we want to transition to the query_repeat_story dialog immediately after the story ends.

<?xml version="1.0"?>
<vxml version="2.1"
 xmlns="http://www.w3.org/2001/vxml">
   <property name="timeout" value="1s"/>

   <form id="read_story">

      <field name="cmd">
         <!-- when story is finished, timeout immediately -->
         <property name="timeout" value="0s"/>
         <!-- only respond when the user says stop -->
         <property name="bargeintype" value="hotword"/>
         
         <prompt>
            <!-- play a story. typically this story would 
                 come from a dynamic feed -->
         	On Wall Street today, 
         	tech stocks rallied with the naz dack climbing 200 points. 
         	Telecommunication giant A T and T 
         	gained a collosal thirty points while software 
         	maker Microsoft gained a strong twenty points 
         	thanks to strong quarterly earnings.
         	<break/>
         </prompt>
         
         <grammar mode="voice"
         root="root_rule"
         tag-format="semantics/1.0"
         type="application/srgs+xml"
         version="1.0"
         xml:lang="en-US">
                  <rule id="root_rule" scope="public">
                           <one-of>
                                    <item>
                                             <one-of>
                                                      <item>
                                                               stop
                                                      </item>
                                             </one-of>
                                             <tag>out.cmd = "cmd_stop";</tag>
                                    </item>
                           </one-of>
                  </rule>

         </grammar>

   
         <noinput><goto next="#query_repeat_story"/></noinput>
   
         <filled><goto next="#query_repeat_story"/></filled>

      </field>
   </form>

   <form id="query_repeat_story">
      <field type="boolean" name="yesno">

         <prompt>Do you want to hear this story again?</prompt>

         <catch event="noinput nomatch">
            Sorry. I didn't get that. 
            Please say yes to hear the story again.
            Say no to return to the main menu.
         </catch>
      </field>

      <filled>
         <if cond="yesno">
            <goto next="#read_story"/>
         <else/>
            <exit />
         </if>
      </filled>
   </form>
</vxml>


[24]7 Inc.| Terms of Service| Privacy Policy| General Disclaimers