Design And Implementation Of Intelligent Voice Dialogue Application Based On Voice Noise Reduction

Smart voice dialogue APP is a mobile phone application that can talk to users with smart voice.In recent years,with the rapid development of information technology,a variety of intelligent products emerge in an endless stream.People's convenient and fast modern life cannot do without a variety of intelligent products.Smart phones are the most common intelligent products.In recent years,the holding rate of smart phones is growing rapidly.Mobile phones help people deal with the big and small things in life,and people use mobile phones more and more frequently.At present,the input of mobile phone mainly relies on two hand typing,and the input channel is single.While driving,walking and other situations,two hand typing also brings security risks to users.At the same time,keyboard typing also brings a lot of inconvenience to users with poor eyesight.In order to solve the above problems,this paper studies the voice noise reduction algorithm and Android application development technology,improves the GRU model for real-time voice noise reduction,establishes a complete corpus and designs a personification interface,and designs and implements an intelligent voice dialogue APP.The APP can interact with the user smoothly by voice,so that the user's instructions are executed when they are spoken,which liberates the user's hands and brings convenience to the user's life.The main contents of this paper are as follows1)The input features of GRU neural network are improved to improve the speed of speech denoising.According to the frequency domain distribution principle of speech signal,the input features of GRU model are simplified.The experimental results show that the improved GRU module can meet the real-time requirements of human-computer voice dialogue while ensuring the noise reduction performance of the algorithm.2)The internal structure of GRU neural network is improved to improve the ability of speech noise reduction.According to the principle of neural network algorithm,the noise parameter fitting part of GRU model is enriched.The experimental results show that the noise reduction ability of the improved GRU model can meet the requirements of subsequent natural language processing.3)Build the natural language processing and multi round dialogue management model.A complete corpus is established,appropriate lexical analysis module is designed,and appropriate Chinese dialogue components are selected to realize daily chat,weather query,music playing and other functions.4)Design and implement intelligent voice dialogue APP.Based on Android studio platform,the intelligent voice dialogue application is developed,and the anthropomorphic human-computer interaction interface is designed.The improved voice noise reduction algorithm,voice text conversion and information interaction with the computer dialogue management module are implemented on the APP with code.