Electronic device and method for interactive speech recognition

Abstract

The invention provides an electronic device for interactive speech recognition. A speech base is stored in the electronic device, and response texts corresponding to different speech texts are recorded in the speech base. A method for the interactive speech recognition comprises the following steps: the electronic device obtaining a speech signal input by a user; transforming the received speech signal to a text message; determining whether a speech text corresponding to the text message exists in the speech base, if the speech text exists, making the speech text a recognition text; if the speech text does not exist, obtaining from the speech base a pre-set number of speech texts which are similar to the speech text, then transforming the pre-set number of speech texts to speech messages in turn and outputting the speech messages, choosing the speech text as the recognition text according to input of the user; and finally, obtaining from the speech base a response text corresponding to the recognition text , and transforming the response text to the speech message and output the speech message. The invention also provides a method for the interactive speech recognition applied in the electronic device. With the electronic device and the method, speech recognition efficiency of the electronic device can be improved.

Description

technical field [0001] The invention relates to an interactive voice recognition electronic device and method. Background technique [0002] The current speech recognition technology mainly pre-establishes a speech database, and records corresponding answers to various questions in the speech database. When the electronic device receives the voice input by the user and converts it into text information, it searches the voice database for a question similar to the text information, and when found, converts the answer to the question in the voice database into voice and outputs it , but due to the different dialects and pronunciations of each user when inputting voice, the problem identified by the electronic device is wrong, and the problem cannot be found in the voice library, so that the electronic device cannot interact with the user. Contents of the invention [0003] In view of this, it is necessary to provide an interactive voice recognition electronic device and met...

AI Technical Summary

Problems solved by technology

When the electronic device receives the voice input by the user and converts it into text information, it searches the voice database for a question similar to the text information, and when found, converts the answer to the question in the voice database into voice and outputs it , but because each user has a different dialect and pronunciation when inputting voice, the problem identified by the electronic device is wrong, and the problem cannot be found in the voice library, so that the electronic device cannot interact with the user

Images