1231 results about "Voice frequency" patented technology

The embodiment of the invention provides a method for full frequency domain digital hearing aid, which comprises the following steps: firstly, acquiring input voice signals of front and back two microphones and performing framing, Fourier transformation and voice scene type recognition; secondly, when voice is mixed with noises, performing noise detection of subframe voice frequency domain signals, beamforming of the two microphones, wind noise processing and inhibition of other noises, compacting the dynamic ranges of frequency domains and inhibiting acoustic feedback; and finally performing the Fourier transformation and overlap-add to obtain output voice signals. The embodiment of the invention also discloses equipment for full frequency domain digital hearing aid. Through the proposal provided by the embodiment of the invention, the problem that the prior digital hearing aid focuses on solving only one aspect of hearing disorder rather than comprehensively take all factors influencing use effect into consideration is solved. Meanwhile, the embodiment of the invention provides a proposal for full frequency domain digital hearing aid. The method, the equipment and proposal have the advantages of quick processing, less resource occupation, low energy consumption and the like.

The invention relates to the technical field of information labeling, and discloses an automatic audio/video label labeling method and system. The method includes the steps that knowledge points and vocabularies of all subjects are captured in advance, and a corresponding subject mapping knowledge domain is built; the subject vocabularies serve as a hot word source, and voice frequencies extracted from an audio/video to be labeled are converted into a text; key words in the text are extracted, and the subject to which the audio/video belongs and the knowledge points are determined according to the incidence relation of the keywords and the mapping knowledge domain; a label corresponding to the audio/video is built, and the label comprises the keywords and the subject to which the audio/video belongs and the knowledge points. By means of the automatic audio/video label labeling method and system, the audio/video source contents can be fully excavated, automatic labeling of labels is achieved, the amount of manual intervention is reduced, and meanwhile, a good basis is provided for the later resource push and other services.

The invention discloses a text-dependent speaker recognition method based on joint deep learning, and belongs to the field of intelligent voice. The text-dependent speaker recognition method comprises the steps that firstly, an FBANK coefficient is extracted and acquired from voice frequency to be tested, the FBANK coefficient is input into a neural network to be calculated after frame extension, and the j-vector of the voice frequency to be tested is acquired; an LDA model is retrained, a predictive threshold value is acquired, finally, the j-vector of the registered voice frequency to be tested of a speaker is normalized with the j-vector of the tested voice frequency to be tested of the speaker, then the LDA model with the predictive threshold value is input, and a predicting result is acquired. The text-dependent speaker recognition method based on joint deep learning has the advantage that the accuracy of the text-dependent speaker recognition can be improved greatly.

The invention discloses a method and a device for realizing e-book annotation on an electronic reader. The method comprises the following steps of: displaying multimedia annotation content in a multimedia annotation library; and responding an annotation inserting request of a user and inserting the multimedia annotation content confirmed by the user into an annotating position selected by the user in an e-book. The method and the device of the invention can provide the multimedia annotation library which comprises user frequently-used marking tools, such as icons, animations, text annotating frames, voice frequency, and the like; a user can call the multimedia annotation library for inserting the multimedia annotation content when reading the e-book; and the electronic reader is used for automatically storing the e-book annotated by the user into a special e-book with user individuality. The invention can meet the requirement that a paper book reading annotation inhibit of the user extends to an electronic reading device and more favorably realize the user individual annotation inhibit in a multimedia way, has convenience, easy use and stability and can popularize the electronic reading device for being universal in an individual user.

The invention provides a system for preventing phone and short message frauds. The system comprises a mobile client side and a server and is characterized in that the server comprises a recording fraud detection module, a fraud identity retrieval module, a speech recognition module, an intelligent analysis module, a voice frequency and fingerprint database, a fraud criminal vocal print database and a fraud information database. According to the system and method for preventing the phone and short message frauds, the technologies of voice frequency and fingerprints, vocal print authentication, speech recognition, natural language comprehension and the like are comprehensively utilized, comprehensive intelligent speech analysis is provided, speech is analyzed from the aspects of incoming call numbers, speech content, incoming call identities, recording data sources and the like, fraud prevention detection is provided, calls/short messages which are suspected by a user are detected comprehensively, suspicious calls are detected in time, and the user is reminded and effectively prevented from being deceived.

The invention discloses a method and device for increasing the volume of received voice and a mobile terminal and belongs to the technical field of noise processing. The method comprises the steps that a first voice signal is collected through a first microphone and a second voice signal is collected through a second microphone; the first voice signal and the second voice signal are processed, so that a user voice signal is obtained through filtration; according to the user voice signal, the intensity of a background noise signal is estimated; according to the intensity of the background noise signal, a voice frequency receiving frequency response curve is determined. By the adoption of the method and device, on the basis of the double-microphone noise reduction technology, the intensity of the background noise signal is estimated, the proper voice frequency receiving frequency response curve is selected according to the intensity of the background noise signal so that the volume of voice of an opposite side can be increased, a user can hear the voice of the opposite side clearly in a noisy environment, and the communication satisfaction degree can be improved.

The invention discloses a method and device for automatically generating background music. The method comprises the steps that relevant data of images or video are extracted; an image or video data stream is generated according to the relevant data; a voice frequency which is matched with the content in the images or the video is searched for in a voice frequency data base according to the image or video data stream. The method and device for automatically generating the background music can automatically generate the music matched with the images or the video, a user does not need to participate in the process, easiness and convenience are brought to the user and the user experience is greatly improved.

The invention relates to a speaker identification method based on neutralization and sound-groove model conversion, the steps comprises (1) extracting voice feature, firstly conducting voice frequency pre-treating which is divided into three parts of sample-taking quantification, zero drift elimination, then extracting reverse spectrum signature MFCC, (2) building emotion model library, conducting Gaussian compound model training, training neutral model according to the neutral voice training of the users, conducting neutralization-emotion model conversion and obtaining emotion voice model by algorithm approach of neutralization-emotion voice conversion and (3) scoring for the voice test to identify the speakers. The invention has the advantages that the technique uses the algorithm approach of neutralization-emotion model conversion to increase the identification rate of the emotive speaker identifying. The technique trains out emotion voice model of the users according to the neutralization voice model of the users and increases the identification rate of the system.

The invention provides a large cluster mechanical three-dimensional garage guiding system and a guiding method, and provides a guiding method and an intelligent and automatic guiding system integrating user identification, display guide, voice frequency guide, vehicle induction, vehicle detection, image acquisition and the like. The purposes of omitting a barrier system on a passageway of a traditional parking lot, greatly increasing vehicle entering and exiting speed and saving management personnel can be achieved.

The invention relates to a multidirectional early-warning positioning and automatic tracking and monitoring device for a monitoring area, which still can monitor a panorama area with a large range even under the condition of low illumination. The monitoring device comprises an early-warning positioning mechanism and a tracking warning mechanism, wherein an early-warning positioning module is arranged inside the early-warning positioning mechanism, and a tracking warning module is arranged inside the tracking warning mechanism. According to the monitoring device, a plurality of directions are subjected to early-warning video collection so as to directly detect the trends of the inside of a plurality of early-warning video images to determine the position information of a monitored moving object, and therefore, the moving object is subjected to tracking video collection, voice frequency collection, illumination warning and sound warning. Meanwhile, an alarm signal is added in a corresponding video image, master control equipment is controlled by passing an instruction back to generate an alarm action, and finally, a plurality of early-warning videos and tracking videos are stored and transmitted to display equipment. The monitoring device has a multidirectional monitoring area regulation function, and the applicability and the flexibility of the monitoring device are obviously improved.

The invention discloses a network violent video identification method based on multiple examples and multiple characteristics. The method for identifying the network violent videos comprises the steps of grasping violent videos, non-violent videos, comments on the violent videos, comments on the non-violent videos, brief introductions of the violent videos and brief inductions of the non-violent videos from a video sharing network, and structuring a video data training set; extracting textural characteristics from textural information of the training set, forming textural characteristic vectors to train a textural pre-classifier, and screening out candidate violent videos by using the pre-classifier; using a shot segmentation algorithm based on a self-adapting dual threshold for conducting segmentation on video segments of the candidate violent videos, extracting related visual characteristics and voice frequency characteristics of each scene to express the scene, taking each scene as an example of multi-example study, and taking video segments as a package; and using an MILES algorithm for converting the package into a single example, using a characteristic vector for training a classifier model, and using the classifier model for conducting classification on the candidate violent videos. By the utilization of the network violence video identification method, bad influences that the network violent videos are broadcasted without constrain are largely lightened.

The invention discloses a method and a device of noise reduction of an earphone. The method of the noise reduction of the earphone includes that a mobile terminal obtains environment noise; the mobile terminal generates a signal with the same amplitude and the opposite phase of the noise; and the mobile terminal outputs the signal through the earphone when outputs voice frequency. According to the method and the device of the noise reduction of the earphone, a general earphone can achieve an effect of the noise reduction without being added with a noise reduction module.

The invention discloses an autonomous fitting hearing aid system based on an intelligent terminal. The system comprises an ear hearing aid used for realizing hearing compensation and a mobile intelligent terminal used for realizing audiometry and fitting. The ear hearing aid and the mobile intelligent terminal carry out wireless communication. In a fitting method, through a questionnaire and audiometry of left and right ears respectively, a hearing condition of a user is acquired. Based on the hearing condition, a hearing map and hearing damage state data are made; hearing compensation prescriptions of left and right ears of the user are generated so as to acquire different voice frequencies of the left and right ears of the user and different sound strength gains, parameters are arranged on hardware of the ear hearing aid, and then a voice signal received by a microphone is segmented according to a frequency, and different amplification and compression are adopted for voices with different frequency ranges. A noise suppression algorithm is used to restrain noises generated by an environment and a hearing aid circuit. Signals of each frequency range are synthesized and finally the synthesized voice signal is output through a loudspeaker. By using the system and the method, fitting cost of the hearing aid user can be greatly reduced.

The invention discloses a voice memo prompting method and device. The method comprises the steps of obtaining user voice frequency prompting information recorded by a voice frequency input device or obtaining character prompting information, converting the character prompting information input by the users into the voice frequency prompting information, searching to obtain events needing to be prompted from a voice frequency database according to the voice frequency prompting information, compiling the events into a list according to voice frequency data corresponding to the events needing to be prompted, then storing the list, sending triggered prompting signals after an infrared sensor detects that users are away from the preset range, after the triggered prompting signals are received, reading the voice data in the compiled voice frequency list, and playing the voice frequency data through a voice frequency output device. According to the prompting method and device, the preset prompting events can be effectively and timely prompted, the triggered prompting information can be timely generated for the sudden random events, and use is convenient and intelligent.

The invention is applicable to the technical field of information and provides a voice box control method and system. The method comprises the following steps: a control terminal sends a voice frequency playing instruction to a main control voice box; the main control voice box determines a voice box combination corresponding to the voice frequency playing instruction and transmits the voice frequency playing instruction to the voice box combination, wherein the voice box combination comprises one main voice box and at least one secondary voice box; the voice box combination carries out voice frequency playing according to the voice frequency playing instruction. According to the voice box control method and system, a wireless Mesh network is formed by the voice boxes, and the control terminal is communicated with each voice box in the wireless Mesh network through the main control voice box in the wireless Mesh network, so that each voice box in the wireless Mesh network is controlled to carry out the voice frequency playing; mutual communication is carried out among the voice boxes in the voice frequency combination of the wireless Mesh network, and the synchronized audio playing is guaranteed, and furthermore, the voice box control flexibility is greatly improved.

The invention relates to a high performance voice frequency amplifying circuit comprising a first communication channel, a second communication channel, a first digital audio signal and a second digital audio signal; the first digital audio signal and the second digital audio signal are generated from a common audio input signal but differ from each other on the signal amplifying rate; the invention also provides a method to amplify a common audio input signal with different amplifying rates so as to further provide the first digital audio signal and the second digital audio signal which differ from each other on the signal amplifying rate. The voice frequency amplifying circuit is especially suitable to be equipped to receive or process the first and the second digital audio signals and cooperatively works with an integrated digital audio signals controller.

The invention discloses a method used for detecting and suppressing transient noise and relates to the technical field of acoustic signal processing. The method used for detecting and suppressing transient noise solves the technical problems that an existing method is poor in real time performance and poor in suppression effect on transient noise. According to the method, acoustic signals are picked up through a microphone first and then preprocessed, residual steady noise is neglected, preprocessed acoustic signals x(n) are obtained, the short-time spectrum of each frame of x(n) is calculated through fast Fourier transform, transient noise detection is performed on x(n) with a transient noise detection method based on high sampling or auto-spectrum coherence, robust suppression is performed on the transient noise with a protection method based on voice harmonic waves and voice and voice frequency and time frequency continuity, the final transient noise suppression gain of each frame is calculated, and finally all the frames are reconstructed through time domain overlap-add. The method is suitable for voice and acoustic signals and can be applied to both a real-time voice and voice frequency communication system and non-real-time voice and voice frequency signal enhancement.

The invention relates to the field of intelligent control over equipment, in particular to an intelligent alternating current contactor capable of being controlled digitally and dynamically in the whole process. A digital chip is used for realizing closed-loop control over the speed in an attraction process of the alternating current contactor (based on coil current falling edge slope), impact energy of a movable/static iron core and a movable/static contact is controlled directly, and bouncing of the iron core and the contact in the attraction process can be reduced substantially; the attraction and supporting process works in a current closed loop chopping mode, chopping frequency is adjusted automatically, and voice frequency noise is prevented from occurring; meanwhile, attraction and supporting currents can be kept stable in a wide input voltage range, and therefore energy-saving silent closed loop operation can be realized; a disjunction process applies direct current negative voltages to the two ends of a coil after rectification and filtration, so that the coil current is made to descend quickly, an electromagnetic system is demagnetized quickly, and the contact is disjunctive quickly at a moment before zero; the intelligent alternating current contactor also has a self correction function of zero-current disjunction, and therefore the problem that control over zero-current disjunction fails due to disperse movement of the contactor, abrasion of the contact and other factors can be solved.

The invention discloses a recording system and method based on multi-microphone array beam forming. The recording system based on multi-microphone array beam forming comprises multiple microphones used for collecting external voice signals, an analog-digital conversion device, a voice frequency digital signal processing device and an output device in a sequential connection mode. The recording direction can be changed in real time, only voice at a certain angle can be recorded, and the mode that other human voice and background noise excepting the voice at the certain angle are effectively eliminated is set. Meanwhile, it is unnecessary to change the position and the direction of a mobile terminal to deliberately get the mobile terminal close to the needed voice, so that the good effect of recording within 5 meters at the maximum is achieved.

The invention discloses a volume regulation method, a volume regulation device and electronic equipment. The volume regulation method comprises the following steps: collecting a voice message emitted from a user, wherein the voice message at least comprises an average volume value; calculating a difference between the average volume value and a stored average normal volume value to obtain a volume rapid change amount under the sub-consciousness of the user; and if the volume rapid change amount is more than or equal to a preset threshold value, controlling to increase the voice frequency output volume. According to the volume regulation method, the volume can be automatically regulated on the basis of the sub-consciousness behavior of the user, and the user experience is improved.

The invention provides a method and device for adjusting voice frequency parameters. The method for adjusting the voice frequency parameters includes the steps that voice frequency application, operating currently, of electronic equipment is obtained; the type of the voice frequency application is recognized, and the voice frequency parameters configured for the type of voice frequency application are obtained; a control instruction is generated responding to the obtained voice frequency parameters; the control instruction is executed, and a voice frequency processor is controlled to adjust the voice frequency parameters. Due to the fact that the types of voice frequency applications are different, the voice frequency parameters are different, and control instructions generated by the electronic equipment are also different, the electronic equipment can adopt different control instructions for controlling the voice frequency processor to adjust the voice frequency parameters so that the voice frequency processor can meet the requirements of different types of voice frequency application.

A hearing aid having a microphone for converting an ambient sound signal to an electric sound signal, a voice detector adapted to determine if a voice is present in said electric sound signal and a frequency analyzer to determine a fundamental frequency of a voice present in said electric sound signal, a signal processor adapted to process said electric sound signal, or an electric signal derived from it, independent of adjustable processing parameters, a memory having corresponding sets of processing parameters and fundamental voice frequencies, a decision unit to select and forward from the memory to the signal processor a set of processing parameters in accordance with a fundamental frequency determined by the frequency analyzer. Further, a method of adapting a hearing aid and to the use of such hearing aid.