| 摘要 |
FIELD: information technology.SUBSTANCE: invention relates to a system for processing user's voice input. Method comprises receiving of the first voice command that includes an indication of the first search query via a communication network from the electronic device associated with the user; receiving of the second voice command from a device via a network from, which includes an indication of the second search query; determining whether the second search query refinement of the first search request, said determining includes: implementation of the first pseudosearch based on a combination of the first and second search queries; performing the second pseudosearch based on the second search query; determination of the relative weighting of the results of the first and second pseudosearch; deciding that the second search query is a refinement of the first search query, if the relative weighting of the results of the first pseudosearch higher than the results of the second pseudosearch.EFFECT: technical result of the invention lies in the effectiveness of voice recognition due to the possibility of transmission of two voice calls, the second request is a refinement of the first request.35 cl, 3 dwg |
