摘要 |
The user interface in an automatic speech recognition (ASR) system is is dynamically controlled, based upon the level of confidence in the results of the ASR process. In one embodiment, the system is arranged to distinguish error prone AS R interpretations from those likely to be correct, using a degree of confidence in the output of the ASR system determined as a function of the difference between the confidence in the "first choice" selected by the ASR system and the confidence i n the "second choice" selected by the ASR system. In this embodiment, the user interfa ce is arranged so that the explicit verification steps taken by the system as a res ult of uncertain information is different from the action taken when the confidence is high. In addition, different treatment can be provided based upon the "consequences" o f misinterpretation as well as the historical performance of the system with respe ct to the specific user whose speech is being processed. In another embodiment, after an ASR system interprets an utterance, the confidence in the interpretation is asse ssed, and three different interactions with the user may then occur.
|