Recognition Mode: Multiple

<listen mode="multiple">

Remarks

Multiple-mode recognition is used in "open-microphone" or dictation scenarios, that is, when the microphone stays on for an extended period of time. Recognitions processed and returned during this period cause an onreco or onnoreco event to occur but do not close the microphone. In this mode, recognition results are returned at intervals until the application makes an explicit stop method call or when the babbletimeout or maxtimeout periods are exceeded. It is important to note that after any onsilence, onreco, or onnoreco event that does not stop recognition, the maxtimeout and babbletimeout periods are restarted. The recoresult property is updated by these events as for the other modes of recognition.

• Case 1 demonstrates that for each phrase recognized, an onspeechdetected and an onreco event occurs and the result is returned. Once the results are returned, the babbletimeout and maxtimeout intervals are restarted.
• Case 2 illustrates how exceeding the babbletimeout interval results in an onnoreco event that ends recognition.
• Case 3 illustrates occurrence of an onnoreco event in response to unrecognizable input. The listen element continues and babbletimeout and maxtimeout intervals are restarted.
• Case 4 demonstrates the occurrence of an onsilence event in response to no input during the initialtimeout interval, and again the listen element continues.

The following event diagram illustrates the multiple-mode recognition time lines.

See Also

listen Element |  Recognition Mode: Multiple Example | Recognition Mode: Automatic | Recognition Mode: Single