453results about How to "Improve realism" patented technology

The invention provides a virtual fitting method and device, multi-target personalized virtual fitting is realized through a multi-target and multi-mode virtual fitting model, and the virtual fitting device comprises a multi-mode feature interlaced learning attention module (MFILA) and a parallel denoising UNet module. Wherein the MFILA module realizes multi-modal condition feature coding through a multi-modal feature fusion module (MFF) and a multi-reference target texture feature extraction module (MRSTF), and the parallel denoising UNet module is mainly responsible for denoising generation. A user uploads a model clothes graph, a target fitting clothes style graph and corresponding text cues, and the system gradually generates a denoised virtual fitting result from pure noise through an MM-VTON model by taking the inputs as conditions. According to the method, the reality sense and the accuracy of virtual fitting are improved by utilizing an advanced image generation technology and multi-modal learning, meanwhile, the demand of computing resources is reduced, and the method has relatively high practical application value and a wide market prospect.

The invention relates to the field of graphic rendering, in particular to a high-efficiency graphic rendering system based on artificial intelligence core computing power, which comprises an intelligent task decoupling module for performing task decoupling on a graphic rendering task; the AI enhanced texture generation optimization module is used for learning a dynamic scene change rule to generate a texture map in real time, reducing calculation overhead in a repeated texture region by using a multiplexing algorithm, and performing super-resolution compensation on low-resolution textures by using a deep convolutional neural network; the AI driving illumination and shadow calculation module is used for predicting an illumination transmission path in a scene by using AI and reconstructing a dynamic shadow effect so as to realize high global illumination and shadow simulation; the task result synchronization module is used for synchronizing rendering data output by each subtask; according to the method, the high-quality texture map is dynamically generated, super-resolution compensation is carried out on the low-resolution texture, repeated calculation is avoided through a texture multiplexing algorithm, the calculation overhead is remarkably reduced, and the real-time rendering efficiency of a large-scale complex scene is improved.

The invention discloses an environment adaptive meta-universe scene perception system and method, and belongs to the technical field of data transmission, and the method specifically comprises the steps: collecting and preprocessing multi-dimensional environment and sensory data in a meta-universe scene in real time through a sensor array, and constructing a full sensory environment data set; radar scanning and depth camera technologies are utilized to accurately obtain three-dimensional information of a physical environment, and a super-reality three-dimensional model is established to simulate a complex ecological system and environment characteristics of the physical world; in combination with real-time full-sensory environment data, an intelligent algorithm is adopted to dynamically adjust meta-universe scene elements, a user behavior prediction technology is introduced, scene layout and an interaction mode are intelligently adjusted according to user history and current states, and user experience is improved; user feedback is collected through VR immersive interaction, a scene perception method is evaluated, and iterative optimization is implemented according to an evaluation result so as to continuously improve the sense of reality and interactivity of the meta-universe scene; and efficient construction and intelligent management of the meta universe scene are realized.

The invention discloses a speech enhancement method and system based on a time-frequency graph convolutional network in the technical field of speech signal processing, and the method comprises the steps: obtaining to-be-enhanced noisy speech data, inputting the to-be-enhanced noisy speech data into an improved metric generative adversarial network (CMGAN), and outputting an enhanced speech time domain signal; the improvement method of the metric generative adversarial network CMGAN comprises the following steps: replacing an encoder module of the metric generative adversarial network CMGAN with a time-frequency graph convolution encoder module; in the time-frequency graph convolution encoder module, a time convolution network and a graph convolution network are used for respectively extracting a long-time dependency relationship and global frequency characteristic information of noise-containing voice data to be enhanced, and dense residual error connection is introduced to fuse multi-scale characteristic information at the same time. The speech enhancement performance is improved under the condition of less model parameter quantity and calculation quantity.

The invention provides a middle and primary school multi-person interactive teaching method and system based on a VR technology, and relates to the technical field of virtual reality, and the method comprises the steps: generating a model through deep learning based on pre-obtained teaching aid model data, determining subcomponents, combining a teaching aid generation engine, and constructing a virtual teaching aid; a color image and a depth image are obtained through a structured light depth camera, two-dimensional pixel coordinates are obtained, depth information is combined, a three-dimensional skeleton curve of limb movement of a student is generated, an optical motion capture result is obtained, an inertial motion capture result is obtained through inertial drift calibration based on the three-dimensional skeleton curve, and a limb movement three-dimensional track of the student is reconstructed; generating a force feedback signal by solving a kinetic equation through a collision detection algorithm based on the three-dimensional trajectory of the limb movement of the trainee; and rendering the virtualization model into a virtual teaching scene, synchronizing the limb movement three-dimensional track and the force feedback signal in real time, and generating a multi-person collaborative interaction process of the virtual teaching aid.

The invention is suitable for the technical field of naked eye 3D, and provides a naked eye 3D interaction method and system based on OSG, and the method comprises the following steps: loading an OSG library and a deep learning model, loading a 3D model, a map and a light source through OSG, and obtaining an OSG scene; collecting interaction data of a user in real time, inputting the interaction data into the deep learning model, and performing real-time analysis; outputting the visual angle parameter and the model parameter obtained through analysis, and adjusting the camera position and the model posture in the OSG scene according to the visual angle parameter and the model parameter; according to the optimized OSG scene, utilizing a computer vision technology to generate a parallax image, and utilizing a deep learning algorithm to optimize the parallax image so as to improve the reality sense and the stereoscopic sense of the 3D effect; and synthesizing the optimized parallax image into a naked eye 3D image. According to the invention, real-time and intelligent adjustment can be carried out according to the actual interaction behavior of the user, and the reality sense and stereoscopic impression of the 3D effect are strong.

The invention relates to the technical field of spinning processing, and particularly discloses a spinning processing gauze tensile strength detection device which comprises a bottom plate, a first fixing plate and two second fixing plates are mounted at the top of the bottom plate, clamping mechanisms are arranged above the two second fixing plates, an automatic feeding mechanism is arranged above the first fixing plate, and a feeding mechanism is arranged above the automatic feeding mechanism. Stand columns are installed at the four corners of the top of the bottom plate, third fixing plates are jointly installed at the top ends of the two matched stand columns, a mounting plate is arranged at the bottoms of the two third fixing plates, and a cutting mechanism is arranged at the bottom of the mounting plate. In the process of detecting the tensile property of the gauze, different conditions of the to-be-detected gauze in the actual use process can be simulated by virtue of mutual cooperation of a plurality of components of the cutting mechanism and the automatic feeding mechanism, so that the tensile property of the gauze can be detected more comprehensively; the credibility and the trueness of the detection of the tensile property of the gauze are improved.

The invention belongs to the technical field of live-action three-dimensional model construction, and particularly relates to a live-action three-dimensional model construction method based on ground multi-source heterogeneous point cloud fusion. The method comprises the steps of unmanned aerial vehicle image acquisition and three-dimensional modeling, ground multi-source heterogeneous point cloud data acquisition, ground multi-source heterogeneous point cloud registration, ground fusion point cloud processing, and unmanned aerial vehicle image three-dimensional model and ground fusion point cloud space registration and fusion modeling. According to the method, the unmanned aerial vehicle image and the ground multi-source heterogeneous point cloud are fused, so that the problem of detail missing in large-range scene modeling is solved. The high precision of the ground point cloud and the wide coverage of the unmanned aerial vehicle image are combined, and the integrity and fineness of the model are significantly improved. In addition, the point cloud color endowing technology enhances the sense of reality of the model, and provides efficient and reliable technical support for the fields of smart cities, cultural heritage protection and the like.

The invention provides a digital human control method and device based on multiple modes and electronic equipment, and relates to the technical field of data processing. In the method, voice data, facial expression data and body movement data of a user are acquired, and the user interacts with a digital human; performing semantic analysis on the voice data by adopting a Transform model to obtain an emotional state and a language habit corresponding to the user; performing feature extraction on the facial expression data by adopting a convolutional neural network model to obtain a facial expression type corresponding to the user; performing action analysis on the limb action data by adopting a hidden Markov model to obtain effective action features corresponding to the user; generating a fusion result based on the emotional state, the language habit, the facial expression type and the effective action features; and adjusting the multi-modal output corresponding to the digital human according to the fusion result. By implementing the technical scheme provided by the invention, the interaction experience feeling of the user can be improved.

The invention provides a method for dynamically constructing a coal mine geology digital twinborn body based on multi-source heterogeneous information. Comprising the steps of AIGC-assisted geological result reading, end-to-end structured information self-adaptive generation, primary geologic body model construction through a generalized triangular prism algorithm, geometric smooth subdivision based on Kriging interpolation, special geological element dynamic updating and coal mine digital twinborn matrix generation. According to the method, the AIGC is introduced to assist in reading the geological result innovatively from the remarkable characteristics of the data intensive normal form of the geological information, and intelligent generation of structured information matched with an automatic modeling updating algorithm is realized through big data assimilation of multi-source heterogeneous information; a quick response process of information processing, primary geologic body generation, geometric smoothing, abnormal geologic body encryption and global geologic body construction is formed, and a high-precision coal mine geologic digital model is dynamically constructed by utilizing geologic information of different geologic survey stages. And the requirements of high precision, renewability and automation of the coal mine digital twinborn body are met.

The invention discloses a mask generation and image decoupling diffusion three-dimensional medical image generation method, relates to the technical field of image processing and analysis, and solves the problems of poor generalization ability, low calculation efficiency and high dependence on training data during image conversion in the prior art. The method comprises the following steps: acquiring a mask image, and processing the mask image by using a multi-condition diffusion probability model to obtain a multi-label mask sequence; screening the three-dimensional medical image data according to a screening condition, determining an informed slice set, and inputting the multi-label mask sequence and the informed slice set into a volume diffusion generation model to obtain a three-dimensional composite image; random noise is added to the three-dimensional composite image, a diffusion semantic refinement generation model is utilized, the three-dimensional composite image added with the random noise is refined, and a refined three-dimensional image is obtained; according to the invention, data decoupling and model generation volume consistency are realized, and a three-dimensional medical image with higher quality can be generated.

The invention discloses a voice cloning method, a voice cloning system, a storage medium and a program product, and relates to the field of artificial intelligence, and the method comprises the steps: collecting the original voice of a target speaker; performing preprocessing operation on the original voice to obtain optimized voice data; extracting voice feature data from the optimized voice data; training the voice feature data through a preset deep learning model to obtain a trained voice clone model; based on the voice clone model, generating clone voice according to the input text content; calculating the similarity between the cloned voice and the original voice; and when the similarity exceeds a preset similarity threshold, outputting the cloned voice. By implementing the method, the accuracy of voice cloning is improved.

The invention provides a heart point cloud reconstruction method and system for echocardiography, and the method comprises the following steps: obtaining three-dimensional heart structure data, and carrying out the preprocessing of the data; reading voxel data of the three-dimensional heart structure data, and performing point clouding to form voxel reconstruction point cloud data; performing standardization processing on the voxel reconstruction point cloud data to obtain all-center point cloud data, and performing segmentation acquisition to obtain standard section point cloud data; establishing a scanning guide network based on the heart structure data to be processed and the standard section point cloud data; and obtaining a test sample, inputting the test section point cloud data in the test sample into the scanning guide network, and outputting the three-dimensional heart structure data of the test sample and the heart point cloud data of the section where the test section point cloud data is located. After the scheme is adopted, an echocardiogram scanning guide network based on the point cloud is provided, a complete three-dimensional heart model is constructed, and meanwhile, the tangent plane where the two-dimensional echocardiogram is located is determined in the three-dimensional heart model.

The invention discloses a monocular video face reconstruction and driving method based on Gaussian splashing and a 3DMM model. The monocular video face reconstruction and driving method comprises the following steps: S1, converting each frame of a face RGB video into an explicit expression face grid based on the 3DMM model; s2, constructing a geometric enhancement module, and carrying out geometric alignment on eye and tooth regions by adopting an improved iterative nearest point algorithm; s3, initializing a 3D Gaussian ball, and driving the dynamic change of the 3D Gaussian ball according to the point change of the face grid under different expressions; s4, based on the characteristics of different face areas, adaptive optimization learning strategy coefficients are distributed; and S5, increasing confidence information for each 3D Gaussian ball, generating a confidence image through rendering, multiplying the confidence image by a rendered image, calculating high-dimensional feature similarity with the input image, and optimizing attribute parameters and rendering effects of the 3D Gaussian balls. According to the method, Gaussian splashing and a 3DMM model are combined, and high-precision reconstruction and driving of a monocular video face are realized.

The invention provides a spacecraft-oriented multi-task data set construction method. A high-precision satellite geometric model is created. The image rendered based on the PBR not only has a bright-eye visual effect, but also conforms to a physical optics principle, and particularly, a normal mapping simulates a star body wrinkle effect. A space simulation scene is constructed by using an unreal engine 4, and a vivid rendering effect is provided. A high-quality satellite image is generated by combining a scene capture component and a target rendering function of an unreal engine, and a semantic segmentation tag, an instance segmentation tag, a depth tag, a normal tag and a pose tag are automatically generated based on a 3D model 2D image automatic labeling method. And a large-scale data set can be quickly generated in a frame-by-frame rendering and automatic labeling mode. Semantic segmentation is carried out on the generated image through the deep learning model DeepLab V3 +, the accuracy and efficiency of model prediction are further improved, and the rendering trueness is greatly improved through detailed analysis and optimization.

The invention discloses a facial image super-resolution method based on facial structure prior fusion network guidance. The method comprises the following steps: constructing a facial image super-resolution network based on facial structure prior; preprocessing the public data set, and dividing the data set into a training set and a test set; training a face image super-resolution network based on face structure prior guidance; and super-resolution is carried out on the images in the test set. A face image super-resolution model based on a generative adversarial network is constructed, a generator part is composed of a rough super-resolution module, a key point heat map extraction module, a fine super-resolution encoder and a fine super-resolution decoder, and a discriminator is composed of a convolutional layer and a maximum pooling layer. According to the face super-resolution model constructed by the method, the problems of detail missing and low trueness of a result image under a relatively large magnification factor in an existing face super-resolution method are solved, so that a high-resolution face image with high trueness and a maintained structure can be reconstructed by the method.

The invention discloses an intelligent directional and quantitative smell generation method in virtual reality olfaction interaction, and relates to the technical field of virtual reality. The invention provides an intelligent, directional and quantitative smell generation method in order to solve the technical problems that in smell generation work in virtual reality in the prior art, the variety and generation precision of smell are limited, an accurate smell concentration control mechanism is lacked, and interactive linkage with user actions and a virtual environment is lacked. In combination with an odor molecule dynamic combination technology, a deep learning prediction model, closed-loop concentration control and a spatial orientation transmission system, richer, personalized and immersive olfactory interaction experience is provided. Comprising a smell synthesis module, a digital smell control unit, an intelligent smell prediction model, a directional smell transmission system and a virtual wind field simulation module. The method is suitable for being applied to the work of intelligent directional and quantitative smell generation in virtual reality olfaction interaction.

The invention is a divisional application of 202311870114.7, and discloses a speech synthesis method, which comprises the following steps: dividing a target text into a plurality of target text segments according to a text sequence; inputting the first target text segment in the plurality of target text segments into a speech synthesis model to obtain a first target audio segment; splicing the first target text segment and the first target audio segment to obtain a text-audio alternating sequence; acquiring a next target text segment, and adding the next target text segment to the text-audio alternating sequence; inputting the text-audio alternating sequence into a speech synthesis model to obtain a next target audio segment of the next target text segment, and adding the next target audio segment into the text-audio alternating sequence until a final target audio segment corresponding to the final target text segment is obtained; and splicing the target audio segments to generate a target audio. According to the invention, text understanding can be carried out by using contexts of texts and audios, and rhythms, melodies and emotions in a scene can be captured.

The invention relates to the field of image data processing, in particular to a three-dimensional image safety processing system and method based on a diffusion model. The system comprises an image format conversion module, an image feature training module, an image three-dimensional structure analysis module, an image light effect simulation module, an image rendering module and an artificial intelligence security and image anti-counterfeiting module. Through the convolutional neural network, deep semantic analysis of image content can be realized by processing a large amount of three-dimensional image data, so that the model not only identifies key structure elements in an image, but also can predict positions and interrelationships of the elements in a three-dimensional space, thereby remarkably improving identification accuracy and application efficiency of the model, and improving the application range of the model. A light tracing technology and a neural renderer of a neural network are combined, and the sense of reality of the rendering process is enhanced by simulating the physical propagation process of light. And meanwhile, the accuracy, authenticity and safety of image processing are effectively integrated, and the reliability and the expression effect in image generation and use are comprehensively improved.

The invention is a divisional application of 202311870114.7, and discloses a speech synthesis method, which comprises the following steps: dividing a target text into a plurality of target text segments according to a text sequence; inputting a final target text segment in the plurality of target text segments into the speech synthesis model to obtain a final target audio segment; splicing the final target text segment and the final target audio segment to obtain a text-audio alternating sequence; obtaining a previous target text segment, and adding the previous target text segment to the text-audio alternating sequence; inputting the text-audio alternating sequence into a speech synthesis model to obtain a previous target audio segment of a previous target text segment, and adding the previous target audio segment to the tail of the text-audio alternating sequence until a first target audio segment corresponding to the first target text segment is obtained; and splicing the target audio segments to generate a target audio. According to the invention, text understanding can be carried out by using contexts of texts and audios, and rhythms, melodies and emotions in a scene can be captured.

The invention provides a construction method and device for a digital twin model of a power distribution network. The construction method comprises the following steps: constructing a three-dimensional live-action model of a power distribution network ground environment in an unreal engine by using an oblique photography technology; constructing a three-dimensional BIM model of the power distribution network underground cable and a channel of the power distribution network underground cable in the three-dimensional live-action model by using a parametric modeling technology; and performing simulation processing on the three-dimensional BIM model by using an unreal engine to generate the digital twin model of the power distribution network. According to the invention, the problem that the three-dimensional space relationship between the cable and the surrounding environment cannot be visually displayed in the prior art is solved.

The invention discloses a digital human personalized shaping method and device based on user psychology and a medium. The method comprises the following steps: acquiring multi-modal data when a user interacts with a digital human; based on the multi-modal data, according to the style preference of the user, establishing a psychological feature model through an adaptive learning mode; dialogue voice of the digital person is generated according to the psychological feature model, and matched facial expressions and limb actions of the digital person are generated according to the dialogue voice and the psychological features of the user. According to the digital human personalized shaping method and device based on the psychology of the user and the medium, the psychological feature model is established based on the style preference of the user, and the interaction content matched with the conversation voice, the facial expression and the body movement is generated, so that the digital human can show more real and natural personalized behaviors, and the user experience is improved. The problem that it is difficult to carry out personalized behavior shaping on the digital human according to psychological characteristics and interaction preferences of the user can be solved.

The invention provides a real-time rendering method for an augmented reality game, and relates to the field of three-dimensional reconstruction, and the method comprises the steps: obtaining the rendering scene description of a game object in a three-dimensional model; generating a rendering fusion feature of the game object according to the rendering scene description, and performing subdivision detection on the rendering fusion feature to obtain a texture element index of the game object in the virtual object model; determining a target scene rendering image of the game object in the rendering coloring model, and determining a rendering embedding identifier corresponding to the target rendering range according to the target scene rendering image; determining a rendering structure level of the game object in the target rendering area according to the reference pose information, and determining a rendering interaction attribute of the game object during rendering task distribution according to the rendering structure level and the texture element index; the game object is rendered in real time according to the rendering embedded identifier and the rendering interaction attribute, and the game scene is generated, and the real-time rendering performance of the augmented reality game can be improved under the influence of visual dislocation generated by the compatible rendering effect.

The invention discloses a multi-modal adaptive digital human system based on deep learning, and the system comprises a facial expression extraction module which extracts facial information from a video image through employing a convolutional neural network in deep learning; the speaking style modeling module is used for analyzing and understanding the language mode of the user by using a long-term short-term memory network so as to predict the intention and emotional state of the user; and the personalized recommendation module is combined with collaborative filtering and a content-based recommendation algorithm to deeply mine and learn historical data of a user, and the invention relates to the technical field of a multi-modal adaptive digital human system. The multi-modal adaptive digital human system based on deep learning combines rich sensor information and advanced artificial intelligence technology, has high-precision emotion recognition ability and flexible adaptive dialogue ability, can effectively increase the sense of reality and intimacy in the process of human-computer interaction, and improves the accuracy of human-computer interaction. And the use satisfaction of the user is greatly improved.

The invention relates to the field of safety monitoring, in particular to an intelligent safety system and method based on operation guiding and monitoring of double unmanned aerial vehicles. The system comprises a front guiding unmanned aerial vehicle which is used for guiding an operator to go to next to-be-operated equipment according to preset operation steps and guiding the operator to follow by using a visual or sound signal; and the rear monitoring unmanned aerial vehicle is used for keeping a certain distance behind the operator, monitoring the position and behavior of the operator in real time, and triggering an alarm mechanism when the operator exceeds a preset guide range. According to the invention, the operator is guided to go to the next to-be-operated device through the front guiding unmanned aerial vehicle, the operator is guided to follow, the time waste of the operator for searching the target device is effectively reduced, and the working efficiency is improved. In addition, the real-time monitoring function of the rear monitoring unmanned aerial vehicle ensures that the operator is always on a safe and correct path, and extra time and resource consumption caused by deviation from a predetermined path is reduced.

The invention discloses a simulation training platform for aircraft flight performance simulation, and relates to the technical field of simulation training, and the platform comprises a data obtaining module which is used for obtaining verification data; the parameter resolving module is used for acquiring structural data of the flight performance simulation model and resolving target model parameters of the flight performance simulation model based on the verification data and the structural data; and the parameter correction module is used for correcting actual model parameters of the flight performance simulation model, the data acquisition module acquires operation data of a pilot after the actual model parameters are corrected, and the corrected flight performance simulation model performs flight performance simulation according to the operation data. According to the method, the target model parameters for constructing the flight performance simulation model are calculated by using the verification data, and the actual model parameters of the flight performance simulation model are adjusted according to the target model parameters, so that the inherent error of the flight performance simulation model is as small as possible, and the fidelity of flight performance simulation is improved.

The present invention discloses a method and system for replicating digital human images based on NeRF, and relates to the technical field of digital human image replicating. The method comprises: collecting digital human images to obtain training data sets; expanding and generating training samples; reconstructing a model using the expanded digital human image data training data; extracting features from the results of digital human image replicating output by the autoencoder, and inputting the feature vector after feature extraction into the classifier model to perform digital human image data screening. The present invention improves the details and authenticity of the generated samples, makes the expanded data closer to the real collected images in terms of fidelity, enhances the adaptability and stability when screening digital human image data, makes the screening model more adaptable when facing diversified replica data, improves the classification effect, and finally ensures the high quality and consistency of the replica data.

The invention discloses a piano partner training device based on an AI technology, which comprises an AI intelligent processing system, the AI intelligent processing system comprises a sound recognition system, a fingering collection system and a dynamic collection system, the sound recognition system is electrically connected with a high-quality microphone mounted on a piano, and the high-quality microphone is electrically connected with the dynamic collection system. The fingering collection system is electrically connected with a plurality of groups of action collection sensors which are mounted on piano keys and respectively correspond to the keys, and the dynamic collection system is electrically connected with a high-resolution dynamic collection camera mounted on the piano; the AI intelligent processing system further comprises a personalized learning module, a diversified practice module, an interactive feedback and guidance module and a cloud updating and learning module. The invention relates to the technical field of musical instrument training, in particular to a piano partner training device based on the AI technology, which can provide powerful computing power and learning algorithms, provide personalized and real-time feedback and guidance for piano learners and further improve the training effect of the piano learners.

The invention provides a road network trajectory generation method and system based on pre-training, and a medium. The method comprises the following steps: S1, carrying out pre-training based on simulation data to obtain a pre-trained autoregression Transform model; step S2, based on real data, carrying out fine adjustment on the auto-regression Transform model after pre-training; and S3, generating a road edge trajectory through autoregressive reasoning by the finely adjusted model. According to the invention, a large amount of data generated by a simulator is used for training and generating a model, and then the model is aligned with observation results of the real world. The generative model provides a higher fidelity than a simulator. Different from a model which only depends on real world data, the method is more effectively popularized to the road network of the whole city range. Besides, as long as a simulator running on the urban road network can be accessed in the pre-training step, a trajectory generation model can be pre-trained for the city, and the trajectory generation model is used as a basic model further aligned with a real world observation result.

The invention relates to an adaptive emotion-driven timbre clone character-to-speech method and device. Segmenting the target text according to sentences, extracting semantic features, and splitting each sentence into a target phoneme sequence; the method comprises the following steps: splitting a reference text into reference phoneme sequences, splicing the reference phoneme sequences with each target phoneme sequence to obtain a combined phoneme sequence, processing a reference audio to obtain a reference voice feature and a frequency spectrum, processing based on a semantic feature corresponding to a sentence, the combined phoneme sequence and the reference voice feature to obtain a corresponding voice feature, and extracting a corresponding emotional feature from each sentence semantic feature, and processing based on the speech feature, the emotional feature and the target phoneme sequence corresponding to each sentence and the frequency spectrum to obtain a speech signal corresponding to the target text. And extracting emotional features from the semantic features of the target text, enabling the emotional colors of the voice signals to fit the text, and automatically adjusting the emotional colors of the voice signals based on the mapping relationship between the preset output features and the emotional features.