| 摘要 |
FIELD: information technologies.SUBSTANCE: client device visualises a reflection, which includes at least one multimodal element of reflection, for which input data is accepted by means of visual modality and voice modality. When a client detects a fragment of a user's speech via voice modality, the client sends audio data of the uplink that represents a speech fragment to the voice recognition module. The application server receives the result of voice recognition generated by the voice recognition module and sends a response to the request of a voice event to the client. The response to the voice event request is sent as a response to an asynchronous HTTP-request of a voice event sent before by the client to the application server. Then the client may send another request of the voice event to the application server in response to reception of the response to the voice event request.EFFECT: reduced time for finding the required content.21 cl, 6 dwg |
