Specifies the duration of silence to determine that callers have finished speaking.
|
Value |
Integer: 0–INT_MAX (milliseconds) A value of 0 disables the timer (a zero-length silence period). DEFAULT: 1500 (1.5 seconds) |
|
How to set |
Shown in precedence order (see Rules of parameter precedence): In a VoiceXML application, using a <property> element. In the browser, using the Speech-Incomplete-Timeout header in a RECOGNIZE method. In a parameter grammar. In a grammar using a <meta> tag (see the text on recognition-time parameters in Parameters set in grammar files). In Management Station set on the Nuance recognition service. If not using Management Station, set in a Recognizer configuration file (User-nrsxx.xml). |
|
Usage |
Frequently used for tuning individual grammars, but seldom changed at a system-wide level. |
This parameter is also known as the “end of speech” timeout (or more correctly, the after end of speech timeout).
The recognizer determines that a caller has finished speaking by waiting a specified amount of time when silence occurs at the end of speech. Usual values are approximately 1500 milliseconds (one and a half seconds). This timer can start repeatedly (during a single utterance whenever the caller pauses) without reaching the set value. When the duration reaches the set value without detecting new speech, it triggers the end of speech detection.
For the magic_word and endpointer modes where multiple chunks of speech are collected from an ongoing stream, this parameter defines the maximum silence at the end of any single chunk. Any speech after this point is part of a new chunk.
Shorter settings result in faster response times, but risk interrupting callers before they are done speaking. For monosyllabic and single-word answers which typically do not contain pauses by callers, the incompletetimeout can be set to a small value such as 500 msec. For longer utterances such as digit or alphanumeric strings where callers may pause between digits or groups of digits, set incompletetimeout to a larger value such as 2000 msec.
(For Nuance Recognizer, the smaller value of completetimeout or incompletetimeout triggers the timeout.)
The recognizer does not allow this parameter to be less than swiep_BOS_backoff. If you set a lesser value, the recognizer will ignore your setting and use swiep_EOS_backoff instead. For example, if swiep_EOS_backoff is 1000 and you set incompletetimeout to 775, 1000 is substituted as the value of incompletetimeout.
For applications that use Nuance Recognizer's magic word and selective barge-in features, set this parameter to a small value (for example, 300).
A typical application sets this parameter as a VoiceXML property. Note the difference in the value format (the voice browser would convert this to "800"):
<property name="incompletetimeout" value="800ms"/>
This property generates the following MRCP header:
Speech-Incomplete-Timeout: 800
Sample definition in a grammar file:
<meta name="incompletetimeout" content="800"/>
Related parameters:
