Voice recognition processing and feedback system, voice response method

Abstract

The invention provides a voice recognition processing and feedback system and a voice response method. The method includes the steps of collecting user's voice information, identifying the user's voice information and transforming the identified voice information into text information, generating a text feedback corresponding to the text information, detecting the user's gender and age, acquiring and storing a male speech database, a female voice database and a children speech database, selecting the female speech database to perform the speech synthesis of the text feedback so as to generate a female voice file to reply to the user when the detected user is a male but not a child, selecting the male speech database to perform the speech synthesis of the text feedback so as to generate a male voice file to reply to the user when the detected user is a female but not a child, and selecting the children database to perform the speech synthesis of the text feedback so as to generate a cartoon voice file to reply to the user when the detected user is a child. With the method, different voice responses with different tones can be generated based on a speaker's age and gender.

Description

technical field [0001] The invention relates to speech recognition and speech synthesis, in particular to a speech recognition processing and feedback system and a speech reply method. Background technique [0002] In the existing speech recognition processing and feedback system, it usually includes three important components: speech recognition, speech understanding and speech reply. In the third part of the voice reply module, a single voice tone and indiscriminate reply are used. Basically, it can only be divided into male voice and female voice. The user's needs for different voices can only be selected by setting in advance. It is impossible to meet the user's needs for diverse voice replies, and it is even more impossible to automatically convert the voice replies preferred by the user for the current speaking user. [0003] For traditional speech recognition processing and feedback systems, the operating platform can be various multimedia platforms with network func...

AI Technical Summary

Problems solved by technology

Even if the identity recognition function is added to some voice systems, the final voice reply module only adds personalized services, and does not make corresponding adjustments to the voice and intonation

[0008] 2. Unable to attract children

[0009] For children, the attractiveness of the voice system is more reflected in the interactive experience of listening and speaking. For the current traditional voice system, it cannot meet the requirements of children for diversity and fun.

Therefore, it is impossible to mobilize the interest of this group to use

In a way hindering the potential function of expanding the speech system as an educational method

Images