This event logs the restart of recognition on the next candidate chunk of audio when the Recognizer is set to magic-word or selective-bargein mode.
During multi-chunk recognition, the first chunk begins with a SWIrcst event. If it is rejected (SWIrrcn), subsequent chunks will log SWIrrst.
SWIrrst uses the same tokens as SWIrcst—recognition start.
Related topics
Related topics
