Speaker and Call Characteristic Sensitive Open Voice Search

摘要: Techniques disclosed herein include systems and methods for open-domain voice-enabled searching that is speaker sensitive. using speech information, information associated with a spoken query to enhance open voice search results. This includes integrating textual index support the entire cycle. Given query, system can execute two matching processes simultaneously. text process based on output of recognition, as well characteristics caller or user voicing query. Characteristics feature extraction metadata about call. The clusters callers according these characteristics. use specific modify recognition results, modifying