38 results about "Vocal organ" patented technology

The invention relates to an application in the technical field of mobile payment safety verification, wherein the application is characterized in that the safety verification of the mobile payment process is achieved by the utilization of the voice recognition technology. A brand-new mobile payment safety verification method based on voice recognition is provided according to voiceprint uniqueness resulting from difference among physical structures of human vocal organs. The method supports voiceprint verification relevant to text and voiceprint verification irrelevant to text at the same time and can be effectively combined with the existing mobile payment products. While the payment process is simplified, the safety level of the payment is promoted. The whole system structure is shown in the figure 1.

The invention provides a method for acquiring a vocal organ profile in a medical image. The method comprises the following steps of: for the medical image, performing binaryzation on the lip and a background area in the medical image by using an automatic segmentation threshold of the lip and the background so as to acquire a lip profile; extracting an upper tooth profile, a lower tooth profile, an upper jaw profile and a lower jaw profile in a face range included in the lip profile; for an image area between the upper jaw profile and the lower jaw profile, acquiring a reliable edge point of a tongue profile; and fitting a tongue edge profile by the reliable edge point of the tongue profile. By the method, the vocal head part and the organ area can be automatically segmented from the image background, and the whole process is automatically completed without manual interaction.

The invention discloses individualized voice collection and semantics determination system and method. The individualized voice acquisition and semantics determination system comprises a voice collection module, a voice unit module, a voice database, a semantics database, a voice correspondence semantics database, a semantics normalization module and a voice correspondence standard semantics database, wherein the voice collection module collects voice sound from the vocal organ of a sounder; the voice unit module classifies the voices collected by the voice collection module and establishes voice units by using monosyllables, bisyllables and multisyllables; the voice database is established by using the voice units as basic units; and the voice correspondence semantics database is established by corresponding the voice units of the voice database to the semantics database in one-to-one way. In the invention, the sounder directly determines the correspondence between voices and semantics text words, no requirement is asked for the sounder, the sounder does not need to be trained, the system does not need to be trained neither, and the accuracy is 99.9%.

The invention discloses a user registration and logon method by combining speaker speech identity authentication and account code protection in network games. A client receives and processes a speech signal, and a server carries out speaker speech identity authentication through a speech template. The method adopts speaker text related speech identity authentication related and account code protection, and has an anti-theft function. At the client, the method comprises: dual-cache sound storage scheme: acquiring and storing a sound signal; and end-point detection and feature extraction: detecting end points in the sampled sound signal to determine start/end frames of an effective speech signal, and extracting feature parameter (linear prediction cepstrum factor) of each frame. At the server, the method adopts dynamic programming to compute matching degree of speaker speed parameter and speech template. If the account code is thieved, since the vocal process and vocal organ of an illegal user are different from a registered user, the illegal user can not easily pass through the speech identity authentication. Even if the illegal user logs on by copying the account code and speech parameter, the server can compare with the prestored speech parameter and detect parameter conformity, causing speech identity authentication failure. After having successfully registered the account code, the user has to register a speech code by speaking and repeating the same text content until enough quantity of speech templates are successfully generated. The user needs to speak the speech code to log on. After the speaker speech identity authentication is successful, the server can confirm a user logon success immediately or after the user has input the correct account code; and after the speaker speech identity authentication fails, the server can determine a user logon failure immediately or confirm a user logon success after requiring the user to input the correct account code.

The invention provides a vocal organ visible speech synthesis system which comprises a voice frequency analysis module, a parameter mapping module, an animation drive module and a motion analysis module; wherein, the voice frequency analysis module is used for receiving the input speech signal of a speaker, judging a mute section according to energy information, coding non-mute section of speech and outputting a speech line spectrum pair parameter; the parameter mapping module is used for receiving the speech line spectrum pair parameter transmitted in real time from the voice frequency analysis module, converting the speech line spectrum pair parameter into a model motion parameter by using the trained Gaussian mixture model; the animation drive module is used for receiving the model motion parameter generated in real time by the parameter mapping module, driving the motion of key points of a virtual vocal organ model so as to drive the motion of the whole virtual vocal organ model. According to the vocal organ visible speech synthesis system, the motion of the model is driven by the corresponding motion parameter generated directly by a frequency domain parameter of the input speech, and therefore, the vocal organ visible speech synthesis system has the advantage of being free from limitations of an online database and a physiological model.

The invention discloses an automatic tongue contour extraction method based on nuclear magnetic resonance images. The method is based on an active appearance model (AAM) algorithm, and a magnetic resonance image (MRI) of a vocal organ serves as a data extraction source. The method is characterized by comprising the steps of firstly performing image annotation and data set division of a development set and an evaluation set according to a MRI, and secondly constructing an AAM and performing automatic annotation of a tongue model according to the AAM. Compared with the prior art, the method has the advantages that the tongue contour extraction is achieved with small errors, the effect on survey errors in the later period is small and can be ignored, MRIs to be processed are large in quantity, and the method for extracting tongue contours through automatic annotation based on the AAM can save a large amount of time and human resources. The method has great significance to a series of speech studies.

The invention provides a pronunciation method of a three-dimensional visual Chinese mandarin pronunciation dictionary with pronunciation being rich in emotion expression ability and relates to the technical field of voice visualization, language teaching, vocal organ animation and facial animation. The method produces the vocal organ animation and produces the facial animation with vivid expressions at the same time. The method has the advantages that based on truly-captured motion data, a physiology motion mechanism of vocal organs and a harden markov model, the built vocal organ animation has coordination and consistency related to the facial animation, and coarticulation phenomena in continuous voice animation can be completely described; data driving models are embedded into physiology models by the utilization of advantages of the physiology models and advantages of the data driving models on describing face partial detail features and reality sense aspects, and the facial animation with high reality sense is generated. An objective performance test and a subjective performance test to a system verify the effectiveness of the pronunciation method in the aspect of intelligent auxiliary language teaching.

The invention relates to a vocal cord-larynx ventricle-vocal track linked physical model and a mental pressure detection method. The physical model includes a mechanical equation set for describing a vocal cord motion model, and an aerodynamics equation set for describing pressure drop distribution in a glottis depth direction and a larynx ventricle-false vocal cord-vocal track direction. A physiological parameter estimation algorithm is designed through the established vocal cord-larynx ventricle-vocal track linked physical model, so that a physiological variation mechanism of phonation in a pressure state is researched. Physiological feature parameters of the vocal cords and the larynx ventricle when a speaker phonates in the pressure state are extracted, and a relation from real voice signals to physiological features is established. According to the estimated physiological parameters, variation features of various vocal organs and the flow state of airflow in the vocal organs under the influence of pressure variation factors are obtained, and the variation features are used for detection of the mental pressure. The detection recognition precision and reliability are improved.

The invention relates to the fields of medical treatment and education for disabled people, in particular to a method for judging the degree of hearing impairment of deaf-mute patients. A method for judging the degree of hearing impairment of deaf-mute patients, using a three-dimensional electromagnetic pronunciation instrument to obtain the pronunciation start time of the pronunciation organs and the start time of the pronunciation organs of multiple normal people when they read the test corpus, and the deaf-mute patients to be judged read through the three-dimensional The electromagnetic pronunciation instrument obtains the pronunciation start time of the corresponding pronunciation organ and the pronunciation start time of the pronunciation organ when the deaf-mute patient to be judged reads the test corpus, and compares the two to judge the degree of hearing impairment of the deaf-mute patient.

The invention discloses a vocal organ three-dimensional modeling method which comprises the steps that an outer apparent vocal organ contour is extracted from a visible image, the contour is matched with a vocal organ universal three-dimensional geometrical model, and a reestablished outer apparent vocal organ three-dimensional model is obtained; an oral cavity inner vocal organ contour is extracted from an X-ray image, the contour is matched with the vocal organ universal three-dimensional geometrical model, and a reestablished oral cavity inner vocal organ three-dimensional model is obtained; and through the matching relation of the visible image and the X-ray image, the reestablished outer apparent vocal organ three-dimensional model and the reestablished oral cavity inner vocal organ three-dimensional model are fused, and accordingly an complete vocal organ three-dimensional model is obtained. According to the method, the vocal organ model which can reflect the real shape of a vocal organ from inside to outside during pronunciation by the vocal organ can be obtained accurately and quickly.

The invention provides a method for generating lattice animation of vocal organs. The method comprises the following steps of defining vocal organ movement key points; generating a two-dimensional lattice of a midsagittal plane of each vocal organ; obtaining texture maps of the vocal organs; performing thin plate spline-based lattice movement calculation. The shape of the midsagittal plane of each vocal organ is calculated according to the positions of the vocal organ movement key points, and the movement condition of the whole vocal organs on the midsagittal planes is directly obtained according to the movement of the vocal organ movement key points, and the method has the advantages of simplifying description parameters for vocal organ movement and reducing the difficulty in making the animation of the vocal organs.

The invention discloses a speech enhancement method and system fusing ultrasonic signal features. The method comprises the steps: firstly predefining an ultrasonic signal, and then actively transmitting and receiving the ultrasonic signal through a loudspeaker and a microphone of a device; and performing channel estimation to obtain a channel impact response, and inputting the channel impact response to a neural network to realize speech enhancement, wherein the channel impact response reflects motion features of facial vocal organs when a user speaks, and the motion features serve as supplementary modal information of speech. According to the invention, the speech enhancement task is assisted by making full use of the user voice action characteristics, the speech enhancement effect is improved, and the method has wide application prospects.

The invention, which relates to the field of the computer technology, provides a pronunciation detection method. The method comprises: collecting a speech signal that is generated when a tested personreads a designated content and a vibration intensity at a specific position outside a vocal organ; according to the speech signal, acquiring a vibration intensity corresponding to a to-be-tested pronunciation in the designated content; and on the basis of a relationship between the vibration intensity corresponding to the to-be-tested pronunciation and a threshold, determining whether the to-be-tested pronunciation is correct. According to the invention, on the basis of non-invasive pronunciation detection, the operation becomes simple and the cost is lowered; and the tested person feels safeand comfortable. Besides, the invention also provides a pronunciation detection apparatus and a phonetic-category learning method and system based on the pronunciation detection method.

The invention relates to a pronunciation evaluation method and device, a storage medium and electronic equipment. The method comprises the steps of displaying an example sentence text to a user; collecting a to-be-evaluated audio read by the user based on the example sentence text; generating a vocal organ action video reflecting the action of a vocal organ when the user reads the example sentence text; generating pronunciation evaluation information based on the vocal organ action video and a vocal organ standard action video corresponding to the example sentence text; and displaying the pronunciation evaluation information to the user. The pronunciation of the user can be accurately evaluated, and whether the pronunciation of the user is accurate or not can be visually reflected.

The invention relates to the technical field of voice recognition, and particularly relates to an intelligent recognition method for incomplete voice of elderly people. The intelligent recognition method for incomplete voice of elderly people comprises the following steps: S1, preprocessing voice data, acquiring an original voice signal, and preprocessing the original voice signal, specifically including windowing and framing of the voice signal; detecting voice signal endpoints; performing voice enhancement processing by adopting a signal subspace enhancement algorithm; S2, extracting voice features, and fusing the extracted feature parameters; and S3, establishing a voice acoustic model. According to the intelligent recognition technology for incomplete voice of elderly people, the problems of slight sound amplitude and great influence of environmental noise caused by aging of vocal organs of the elderly people can be reduced, and the voice features fused by adopting the sound parameters can be closer to the voice features of the elderly people, so that data comprehensively representing the voice features of the elderly people can be acquired, and the recognition degree of incomplete voice and fuzzy voice of the elderly people is improved.

The invention relates to a video generation method and device, a storage medium and electronic equipment, and the method comprises the steps: inputting a to-be-processed text into a video feature generation model to obtain a video feature sequence; and generating a vocal organ action video based on the video feature sequence. The vocal organ action video generation efficiency can be improved.

The invention discloses a 3D face animation synthesis method and system, and belongs to the technical field of artificial intelligence, and the method comprises the steps: collecting the audio features of the voice of a user and the rhythm features of a text; taking the audio features and the rhythm features as input of a pre-trained trajectory prediction network to obtain a vocal organ motion trajectory corresponding to the user; and driving the 3D head network model by using the vocal organ movement track of the user to achieve lip sound synchronous 3D face animation synthesis. According to the method and system, the motion trail of the vocal organ is used as the animation parameter of the 3D face model, the 3D face model is driven through the motion trail to achieve the face animation with synchronous lip sound, and the calculation amount is greatly reduced.

The invention discloses a vehicle rearview mirror control method and device, a vehicle and a storage medium, and the method comprises the steps that of obtaining sound source position of a target object, and the sound source position is used for representing the spatial position of a sound production organ making a sound by the target object; determining a binocular position of the target object according to the sound source position; according to the binocular position, adjusting the vehicle rearview mirror to the target position, and therefore the vehicle rearview mirror is adjusted through sound source positioning, and great convenience is brought to vehicle driving of a user.

The invention discloses a pronunciation correction method based on vocal organ form and behavior deviation visualization. The method comprises the steps: performing the foreign language learning modeling for a learner, acquiring the voice content of the learner and the organ motion corresponding to the pronunciation, and then calling a corresponding pronunciation module in a standard library according to the target foreign language selected by the learner; searching the deviations of the lip position, tongue position, tooth position, expiratory volume and jaw height of the learner and the corresponding data in the standard library by comparing the same pronunciation contents and organ movement patterns of the learner and the standard library, and visually feeding back the difference of the organ forms between the learner and the organ of the standard sound and the deviation between the two to the learner, and then updating the basic information of the learner. The method achieves therecording of the learning ability of the learner at any time to provide the learner with a coherent executable way of learning foreign languages.

The invention relates to a rapid dysarthria detection method based on modal decomposition. The method comprises the following steps: collecting an original voice signal; preprocessing the original voice signal; performing acoustic feature extraction on the framed and windowed signal S based on modal decomposition to obtain statistical features; the statistical features are input into a machine learning classifier, dysarthria detection is achieved, and the machine learning classifier is a support vector machine (SVM) model. According to the method, the limitation of traditional acoustic features in a nonlinear time-varying system is overcome, the IMF obtained through decomposition contains time-frequency information of original audio signals on different levels, voice physiological information of dysarthria patients can be well captured, pathological changes of vocal organs are reflected, and the accuracy of the voice physiological information is improved. The accuracy and robustness of dysarthria detection are improved; the method can adapt to non-linear and non-stable voice signals, and further improves the detection effect of dysarthria.

The invention provides a method and system for predicting a human body sound production effect, and relates to the technical field of voice improvement. The method for predicting the human body sound production effect comprises the following steps of acquiring sound production organ information of a user; acquiring corresponding predicted postoperative sound production organ information according to the sound production organ information of the user; and inputting the predicted postoperative sound production organ information into a preset sound production effect model to generate predicted sound production effect information, wherein the preset sound production effect model contains sound production organ information and sound production effects of a plurality of samples, and the predicted sound production effect information can be obtained by comparing the input predicted postoperative sound production organ information with the sound production organ information in the preset sound production effect model; and by predicting the sound production effect information, the user can know the postoperative sound production effect, so that the postoperative effect can be known in advance before the operation, the data support of the operation effect is provided for the user, and the user can conveniently know the operation scheme and effect.