Real-time feedback method and system of long voice continuous recognition and recognition result

Abstract

The invention discloses a real-time feedback method and a system of long voice continuous recognition and a recognition result. The method comprises the following steps: extension possibility from a sentence end point to an identification start point in an identification network is increased; in the process of decoding, when an identification path extends, the identification path can not only generate a single sentence, but also generate a plurality of continuous sentences, and large space video signals are separated through overall optimization of acoustics and linguistic probability; common parts of optimal history path of all active nodes are detected at fixed period, and identification word order which is fixed at the present moment is obtained; updated local identification result is fed back to a user in real time; and decoding space which is corresponding to identification parts and ensured is recycled. When the real-time feedback method and the system of the long voice continuous recognition and the recognition result are used, endpoint detection algorithm is not needed, continuous and sustainable identification of long voice signals are achieved, and human-computer interaction experience is obtained for users.

Description

technical field [0001] The invention relates to the technical field of speech recognition, in particular to a method and system for continuous recognition of long speech and real-time feedback of speech recognition results. Background technique [0002] With the development of the mobile Internet, as one of the most natural and efficient means of human-computer interaction, speech recognition and related technologies are more and more widely used in daily life. Speech recognition is no longer just a beautiful vase. It has actually entered the lives of ordinary people, bringing convenience and happiness to our society. [0003] In practical applications, for the application of real-time speech recognition for large-scale speech signal input, the traditional method is to use the method of online endpoint detection to detect the start point and end point of the speech online, and notify the online decoder to perform Corresponding operation: After finding the starting point of ...

AI Technical Summary

Problems solved by technology

[0009] In view of this, in order to better solve the problem of continuous recognition of long speech signals and at the same time provide users with a better feedback experience, the present invention provides a method and system for continuous recognition of long speech signals and real-time feedback of recognition results. , reduce the dependence on the endpoint detection algorithm, introduce more judgments on semantic integrity, use the global optimal method to find the end point of the sentence in the speech signal, and realize the continuous recognition of large-scale speech; at the same time, real-time detection of the The determined optimal recognition result can be fed back to the user without waiting for the end of the sentence, thus improving the response speed

Images