The Design And Implementation Of Voice Interaction Based VR Image System

The use of immersive virtual reality technology to present video narratives not only enhances content expression,but also provides participants with a good experience,which has gained widespread attention in recent years.Adding interactive behaviors into VR images can effectively improve sense of participation and interest in watching the film,strengthen their understanding of the content,and deepen their memory of the story.At present,the interaction methods used in VR images still have problems such as interference of immersion feeling,lack of emotional resonance,low input efficiency,need to use additional equipment and need a certain learning process.The development of artificial intelligence solves the technical problem of inaccurate dialect recognition and inaccurate semantic analysis in traditional speech interaction,so that its advantages can meet the possibility of realizing various types of interaction in VR images.Based on HTC Vive VR helmet,UE4 engine and IFLYTEK intelligent voice service,this paper develops a voice interactive immersive VR image system,which guides users to participate in interaction and complete viewing by using natural voice.Users are guided to participate in the interaction and watch the film through voice interaction.In this system,users have a natural dialogue with the characters of the system,which triggers the story and can get a more immersive interactive experience;The system will play users’ favorite song,and trigger users’ memories of the past in combination with the scene atmosphere of sunset,so as to obtain emotional resonance;Users can participate in scene activities from the first perspective and feel the visual impact of the scene.Voice interaction also makes it easier for users to use the system,which was verified in the evaluation.In addition,the system solves the technical problems of UE4 voice capture,real-time format conversion of the captured audio in the engine,intelligent response to the voice input by the user,extraction of Title keywords from the voice and audio files input by the user,online access to songs and real-time playback through keywords,real-time voice synthesis of audio data in UE4 engine,dynamic load and so on.