476results about How to "Improve naturalness" patented technology

An ear terminal includes a sealing section arranged for use in the ear meatus of a human, with an inner microphone having a sound inlet for being directed into the meatus and an electronic unit including filtering elements coupled to the inner microphone for filtering the signal from the inner microphone, the filtering elements being programmable to transform the signals based on the sounds received in the ear by the inner microphone into sounds having essentially the characteristics of spoken sounds of the wearer of the ear terminal.

An apparatus for providing voice conversion using temporal dynamic features includes a feature extractor and a transformation element. The feature extractor may be configured to extract dynamic feature vectors from source speech. The transformation element may be in communication with the feature extractor and configured to apply a first conversion function to a signal including the extracted dynamic feature vectors to produce converted dynamic feature vectors. The first conversion function may have been trained using at least dynamic feature data associated with training source speech and training target speech. The transformation element may be further configured to produce converted speech based on an output of applying the first conversion function.

A manually actuated input device for commanding machine- and/or computer-assisted control operations for kinematic motions of a real or virtual multipart object, including a force/moment sensor with which linear displacements in the form of translational movements in the direction of three axes (X, Y, Z), each standing perpendicular on the other, of a three-dimensional rectangular system of coordinates and/or rotational excursions in the form of rotational motions (A, B, C) about these three axes are sensed and converted into commanded motions of the object to be controlled is characterized by the commanded individual linear displacements and/or rotational excursions of the force/moment sensor being assigned specific kinematic motion patterns of parts of the real or virtual object thereby permitting manipulation or animation thereof as a transforming interlink and by the commanded velocities of the corresponding individual linear displacements and/or rotational excursions of the force/moment sensor being additionally weightable as a kinematic interlink assignment. The invention can be put to use e.g. in kinematic animation operations of human-like robots, figures or virtual animate beings.

The invention discloses a psychological stress assessment method based on multi-physiological-parameter integration. The method includes: designing a reasonable stimulation program, acquiring four types of electrophysiological signals, namely electrocardiogram signals, electromyographic signals, pulse wave signals and electroencephalogram signals from people suffering psychological stress; extracting affective features of the four types of electrophysiological signals; subjecting the extracted features to feature selection by means of Relief algorithm, genetic algorithm optimization and the like; acquiring related integration functions on the basis of basic probability assignment mass. According to the method, multi-parameter signals are subjected to acquisition, preprocessed, feature selection and psychological stress affective recognition and are integrated; compared to single-parameter classified recognition or multi-parameter data-level or feature-level integration, the method allows data information to be more fully utilized and psychological stress emotions to be more accurately recognized.

A method for synthesizing voice based on rhythm character includes text processing program formed by text standardizing step, rhythm structure analysis step and language treatment step, synthetic element selecting program formed by element confirming step, matching step, pasting-up step, optimizing and screening step; voice synthesization processing program formed by base frequency outline generating step of phrase unit, base frequency outline generating step of syllable unit and intonation superposing step.

The present invention discloses a children speech emotion recognition method, comprising: creating a training voice database, classifier training and recognizing emotion. The classifier training comprises the step of extracting the speech emotion features, which includes performing pretreatment on the analog speech signal, including sampling, denoising and other basic signal treatments, performingHHT transformation on the pretreated emotion sentences, obtaining the Hilbert spectrum of the emotion speech; performing amplitude feature extraction, intonation feature extraction, tone feature extraction and rhythm feature extraction, and obtaining the emotion feature; reducing dimension of the emotion feature, and obtaining different quaternionic feature of emotion space; performing calculation on the sample set of the training speech database, obtaining the projection of a series of quaternionic features in national emotion state in the emotion space i, thus forming an emotion feature projection set; and carrying out classification training based on the SVM/HMM mixed system by using the data in the emotion feature projection set.

The invention discloses a method for creating a three-dimensional model by using a sketch, which comprises the following steps of: dividing strokes of a freehand sketch into primitive strokes functioning as the outline of an object and threaded strokes acting as three-dimensional effect, identifying the strokes respectively and regularizing the primitive strokes; indicating the geometric constitution of the object by using a conditional rule generation method and realizing rule-driven reasoning by using primitive adjacency relationship under the guidance of drawing threads so as to identify object units and recover the depth information of each object unit; automatically discriminating and completing hidden lines of the object units according to geometric attributes of the object units under the guidance of a drawing thread rule so as to generate a complete three-dimensional object unit; and recreating the three-dimensional model of the object by using a geometric splicing method according to topological relation among the object units. The method has the advantages of expressing the three-dimensional effect with auxiliary threads, such as shade lines and the like, and drawing the perspective projection outline of the object with the freehand sketch so as to improve the degree of freedom of drawing.

A method for processing voice synthetic text based on rhythm structure includes comparing inputted text with preset special symbol table to output legal pronunciation character string, comparing legal pronunciation character string according to participle rule and rhythm structure analysis rule to output labeled character string with rhythm structure information, comparing labeled character string with preset rhythm rule and phonetic table word by work and outputting label phonetic code string labeled rhythm information.

The embodiment of the invention discloses a video processing method and device, terminal equipment and a storage medium. The method comprises the steps of obtaining a to-be-processed video; obtaininga target audio clip and a face image sequence in the to-be-processed video; performing emotion analysis on the face image sequence to obtain emotion features; performing voice analysis on the target audio clip to obtain statement features which are used for representing keywords in the target audio clip; based on the emotional characteristics and the statement characteristics, determining reply contents and expression behaviors of the virtual character; and generating and outputting a reply video for the to-be-processed video, wherein the reply video comprises voice content corresponding to the reply content and a virtual character for executing the expression behavior. According to the embodiment of the invention, the emotional characteristics and the statement characteristics of the character can be obtained according to the speaking video of the character, and the virtual character video matched with the emotional characteristics and the statement characteristics is generated as thereply, so that the sense of reality and naturalness of human-computer interaction are improved.

The present invention discloses a speech enhancing method, a speech enhancing device and a denoising communication headphone. In the solutions of the present invention, a first sound signal that comprises a user's speech signal transmitted through coupling vibration and an ambient noise signal transmitted through the air and a second sound signal that is mainly an ambient noise signal transmitted through the air are picked up by a primary vibration microphone and a secondary vibration microphone, respectively, that have a specific relative positional relationship therebetween, and the ambient noise signals picked up by the two vibration microphones are correlated with each other; a control parameter used to control an updating speed of an adaptive filter is determined according to the first sound signal and the second sound signal; the first sound signal is denoised and filtered according to the second sound signal and the control parameter; and the denoised and filtered speech signal is further denoised and speech high-frequency enhancement is performed thereon. The technical solutions of the present invention can effectively improve the signal to noise ratio (SNR) and the quality of speech in an environment of highly intense noises.

The invention discloses a method for scratching a target object from a digital picture, which belongs to the technical field of the image processing and the human-computer interaction. The method comprises the following steps that: a plurality of nodes are arranged on the boundary of a target object through a human-computer interaction device, a straight line or a curved line is arranged between nodes, and the lines are enclosed to form a boundary outline of the target object; any target point on the node or on the nodal interval line is pressed through the human-computer interaction device, and different operations are respectively executed by judging whether the target point after being pressed is dragged within a given time, and the operations comprise: directly changing the straight or curved shape of the line, changing the nodal position and adjusting the shape of the line which is connected with the nodes, inserting new node, and deleting existed node. With the method and the system, target object can be simply, conveniently, correctly and directly scratched from the digital picture image on a plurality of electronic devices.

The present invention provides a system and method to enhance speech intelligibility and improve the detection rate of automatic speech recognizer in noisy environments. The present invention reduces an acoustically coupled loudspeaker signal from a plurality of microphone signals to enhance a near end user speech signal. A decision unit checks a system configuration parameter to determine if the cleaned speech is intended for human communication and/or Automatic Speech Recognition (ASR). A formant emphasis filer and a spectrum band reconstruction unit are used to further enhance the speech quality and improve the ASR recognition rate. The present invention can also apply to devices which has a foreground microphone(s) and a background microphone(s).

The invention provides a Tibetan-Chinese speech synthesis method and device and aims to synthesize input Chinese or Tibetan statements to be synthesized, through a Chinese-Tibetan hybrid corpus preliminarily established. Chinese or Tibetan speeches can be synthesized at the same time by the method and the device. Compared with a traditional HMM-based (hidden Markov model based) speech synthesis system, the method and the device have the advantages that a speaker adaptive training process is added to a training phase to acquire a Chinese-Tibetan hybrid speech average model, the influence caused by speaker differences in a speech library can be reduced through the speaker adaptive training process, and quality of synthesized speeches is improved; on the basis of the average model, and a Tibetan or Chinese speech excellent in both naturalness and fluency can be obtained through synthesis of few Tibetan or Chinese corpus data through a speaker adaptive conversion algorithm; the research is significant to promoting the development of communication with minorities and the development of minority speech technologies.

The invention provides a three-dimensional human modeling method by using cartographical sketching, comprising the following steps of: 1. analyzing a sketch: decomposing a human sketch into a skeleton stroke set Ss, an outline stroke set Sp and a thread stroke set Sc; 2. extracting the model characters: extracting a human skeleton structure on a predefined human mesh model MH; 3. recovering the depth: identifying a thread determination result on the thread stroke set according to the skeleton structure and calculating the three-dimensional posture according to the human skeleton structure based on the two-dimensional human posture; and 4. reconstructing three dimension: deforming the section outline characteristics of four limbs and a trunk mesh model. By using the cartographical sketching to draw human front elevation skeleton and outline and directly generating the three-dimensional mesh model, the invention accords with the plane drawing habit of an animation role designer, simplifies the human modeling flow and greatly reduces the time for users to construct human models.

Embodiments of the present invention pertain to adaptation of a corpus-driven general-purpose TTS system to at least one specific domain. The domain adaptation is realized by adding a limited amount of domain-specific speech that provides a maximum impact on improved perceived naturalness of speech. An approach for generating optimized script for adaptation is proposed, the core of which is a dynamic programming based algorithm that segments domain-specific corpus into a minimum number of segments that appear in the unit inventory. Increases in perceived naturalness of speech after adaptation are estimated from the generated script without recording speech from it.

The invention relates to a Chinese sign language recognition method based on kinect. The method comprises the following steps of acquiring depth information of hands and the 3D (three-dimensional) coordinate information of bone joint points of main parts of a human body through kinect; respectively acquiring hand characteristic information, position characteristic information and direction characteristic information by processing the characteristics of the acquired information; acquiring a hand element, a position element and a direction element for the acquired hand, position and direction characteristic information by adopting different polymerization algorithm, carrying out the element matching by utilizing a neighboring method and a template matching method adopting an Euclidean distance as a similarity measurement criterion, and determining the hand sign implication. Due to the adoption of the method, a purpose for effectively recognizing the Chinese sign language with vast vocabulary can be realized; and moreover, each element is independently acquired in parallel, so that the Chinese sign language can be recognized in real time. By implementing the method, different sign languages can be recognized in real time, so that Chinese deaf-mute can effectively communicate with other people.

The invention discloses a video processing method, device and system, a terminal device and a storage medium. The method comprises the steps of obtaining the interaction information inputted by a user; identifying the interaction information, and obtaining a reply audio corresponding to the interaction information; inputting the reply audio into an audio visual prediction model to obtain a face model expression parameter corresponding to the reply audio; based on the facial model expression parameters, adjusting a preset facial model to obtain a simulated facial image corresponding to the facial model expression parameters; replacing the preset face image in the preset video with an analog face image, replacing the preset audio with a reply audio, and obtaining a replaced reply video; and outputting a reply video for the interaction information. According to the present invention, when the user converses with a robot, for the conversation content of the user, the robot reply video which simulates a real person and has a natural expression can be provided to interact with the user, the method and the device can be suitable for a customer service robot, the reality and the naturalness of human-computer interaction are improved, and the user experience is optimized.

The invention relates to a text to speech (TTS) method and system. The method comprises: performing text analysis and linguistic analysis on an inputted text so as to generate front-end scripts comprising corresponding speech units; obtaining, verifying and correcting the front-end scripts; and obtaining the corrected front-end scripts for synthesizing correction speech. By adopting the provided TTS method and system, the errors of the front-end scripts such as word-segmentation errors and polyphone phonetic notation errors can be corrected so that the synthesized speech is more understandable and more user friendly for users, a conventional TTS's shortcomings of insufficient prediction accuracy of rhythm can be overcome, and the synthesized speech is more natural and more expressive.

The invention discloses a multipurpose six-legged robot remote control system based on an Xtion device. The system comprises a wireless network module, a motion control unit, an action editing module, an SLAM module, a gesture recognition unit and a speech recognition module. Through the wireless network module, the remote control system controls a six-legged robot. The motion control unit is used to control the operating mode and travel of the six-legged robot. The action editing module is used to edit the movement of the six-legged robot. The gesture recognition unit and the speech recognition unit realize gesture and speech control of the robot respectively. The SLAM module is used to locate the six-legged robot, creates an environment map, and designs two-dimensional and three-dimensional SLAM algorithms to create the environment map. The system provided by the invention has the advantages that the functions of wireless network connection, SLAM, the action editor, gesture recognition and speech recognition are integrated on a six-legged robot platform, so that the six-legged robot can complete tasks better.

In this method, a customer supplies the synthetic information to the synthesis system. It includes the synthetic text information. The synthesis system, by analyzing the text and the rules of the rhyming, combines the synthetic text with audio data. The invention constructs the background audio library, and adds the CSSML mark of the background sound to the synthetic context to strengthen the audio effect.

The invention relates to a voice synthesis method and device, a computer readable medium and electronic equipment. The method comprises the steps that voice feature information of a multilingual textand language feature vectors of all languages in the multilingual text are acquired, and the voice feature information comprises phonemes, tones, segmented words and rhythm boundaries; and voice synthesis is performed according to the voice feature information and the language feature vector to obtain first audio information corresponding to the multilingual text. Therefore, the accuracy and understandability of the first audio data are improved, and a user can quickly understand the text content corresponding to the first audio data. In addition, pause can be carried out at the natural rhythmboundary during speech synthesis, so that the naturalness and fluency of the first audio information can be improved. Besides, the voice synthesis method can realize smooth conversion of different languages, supports voice synthesis of texts of various languages, and does not limit specific languages, namely, has wide applicability.

A device and a method to be used by laryngeally impaired people to improve the naturalness of their speech. An artificial sound creating mechanism which forms a simulated glottal pulse in the vocal tract is utilized. An artificial glottal pulse is compared with the natural spectrum and an inverse filter is generated to provide an output signal which would better reproduce natural sound. A digital signal processor introduces a variation of pitch based on an algorithm developed for this purpose; i.e. creating prosody. The algorithm uses primarily the relative amplitude of the speech signal and the rise and fall rates of the amplitude as a basis for setting the frequency of the speech. The invention also clarifies speech of laryngectomees by sensing the presence of consonants in the speech and appropriately amplifying them with respect to the vowel sounds.

The invention discloses a prosodic hierarchy labeling method. The method comprises the steps as follows: acquiring to-be-labeled text data and audio data which have a corresponding relation; extracting a to-be-labeled text characteristic set of each word according to the to-be-labeled text data; extracting an acoustic characteristic set of each word according to the audio data; acquiring a prosodic hierarchy structure by a prosodic hierarchy labeling model according to a word identification of each word, the to-be-labeled text characteristic set of each word and the acoustic characteristic setof each word. The invention also discloses a model training method, a prosodic hierarchy labeling device and a model training device. The prosodic hierarchy labeling model is established by combination of text characteristics and acoustic characteristics, richer characteristics can be provided for prosodic hierarchy labeling, the prosodic hierarchy labeling accuracy can be improved, and the voicesynthesis effect can be enhanced.

The invention relates to a camera control method. The camera control method includes acquiring eyesight information of users; controlling picturing of cameras according to the eyesight information. Picturing images of the cameras can change as changes of the eyesight information of the users without changing picturing positions or manual operation of the users, the users are greatly facilitated, and nature and friendliness of man-machine interaction are improved. The invention further discloses a device used for implementing the camera control method.