| 主权项 |
1. A method, comprising:
receiving, by one or more processors of a data processing apparatus, a first string related to a natural-language voice user entry and a second string including at least one natural-language refinement to the user entry,
the first string and the second string being a pair of successive user entry strings; parsing, by the one or more processors of the data processing apparatus, the first string into a first set of one or more tokens and the second string into a second set of one or more tokens; determining, by the one or more processors of the data processing apparatus, at least one refining instruction from the second set of one or more tokens; generating, by the one or more processors of the data processing apparatus, a plurality of candidate refined user entries based on the first string, the second string, and the at least one refining instruction,
each of the plurality of candidate refined user entries being generated based on at least a portion of each of the first string and the second string,each of the plurality of candidate refined user entries being different than the first string and the second string; selecting, by the one or more processors of the data processing apparatus, a refined user entry from the plurality of candidate refined user entries; and outputting, by the one or more processors of the data processing apparatus, the refined user entry. |
