32 results about "Voice problem" patented technology

The invention provides a voice problem searching method based on semantic recognition and family education equipment. The method comprises the steps of capturing voice information is transmitted froma user by the family education equipment, performing semantic recognition processing on the voice information, obtaining a semantic content which corresponds with the voice information, determining whether a target waking word which matches the target semantic content exists in a preset waking word set, wherein the waking word set at least comprises keywords with appearance frequency which exceedsa preset frequency threshold in the semantic content of all voice problems which are input by the user through a voice problem searching function before capturing the voice information; and when existence of the target waking word which matches the target semantic content in the waking word set is determined, determining a target problem according to the target semantic content, searching a target answer of the target problem, and outputting the target answer. The voice problem searching method and the family education equipment according to an embodiment of the invention can simplify a problem searching process and furthermore improve problem searching efficiency.

The invention relates to a voice search-based review content recommendation method and an electronic device. The method comprises the steps of identifying target learning content from an acquired voice problem; determining a target attribute of the target learning content, wherein the target attribute comprises a first weight or a second weight, the first weight is used for representing a probability that the target learning content is identified as meaningful content, and the second weight is used for representing information volume of the target learning content. The electronic device generates different review plans according to different target attributes, and recommends corresponding target review content according to time nodes in the review plans, wherein time nodes are set and distributed in the review period according to The Ebbinghaus Forgetting Curve, and number of recommendation times in the review plan is set according to the target attribute of the target learning content. Review content related to voice input problems can be recommended automatically according to a human brain information memory law, so that learning efficiency is improved.

The invention provides an ill-conditioned voice evaluation method based on Chinese voice, which comprises the following steps: completing automatic alignment of input continuous voice signals and corresponding texts by using an automatic Chinese voice recognition system, and completing phoneme segmentation; segmenting the continuous voice signal into vowels, clear consonants and turbid consonantsaccording to the time sequence of phoneme segmentation, and then extracting features; wherein the extracted features form a feature parameter set, and the feature parameter set is screened by a feature selector and then input into a classifier to obtain a fine score of the voice problem. According to the method, a large-scale normal voice database is used for training an automatic voice recognition system; the segmented voice sequences are classified according to the sounding characteristics of different phonemes, the characteristic parameters are designed for different phoneme types, finally,the objective overall score and each detailed score for the voice problem are obtained, and an important reference is provided for clinical diagnosis and rehabilitation treatment.

The invention relates to a voice topic-searching method based on a customized corpus, and family education equipment. The method comprises the steps of: acquiring voice information input by a user, determining personal data of the user, and determining a target corpus matched with the personal data, wherein the personal data at least includes the current age of the user; and the target corpus includes different voice questioning methods of user groups in the age group, which the current age belongs to, for different problems; and then, identifying the voice information according to the targetcorpus so as to extract a target problem included in the voice information, searching a target answer of the target problem, and outputting the target answer. By means of the embodiment of the invention, the identification accuracy rate on the voice problem of a student can be increased; and thus, the topic-searching accuracy rate can be increased.

The present invention relates to the technical field of electronic devices, and especially to a voice topic search method and an electronic device. The method comprises the steps of: obtaining input voice information; analyzing the voice information to obtain a target problem included in the voice information; and obtaining the target result matching with the target problem, and outputting and displaying the target result. The voice topic search method and the electronic device can identify and solve the input voice problem with no need for a manual input problem in an input mode through a keyboard, is simple and convenient to operate, and can improve the learning efficiency when children use the electronic device to perform assistant learning.

The invention relates to and identification method based on voice search and a tutoring machine. The method includes: the tutoring machine detects whether a first voice message input by a user is received or not; if so, the tutoring machine determines a target accent database corresponding to the accent of the user inputting the first voice message from a total accent database acquired in advance,wherein the total accent database is formed by user accents matched with preset statistics dimensionality, and the preset statistics dimensionality comprises areas; the tutoring machine uses the voice content, identified according to the target accent database, of the first voice message as the to-be-searched content of the user. By the method, the voice problem identification accuracy of the tutoring machine can be increased, and the accuracy of the voice problem search results of the tutoring machine is increased.

The invention discloses a gamification language study method and system based on voice control. The method comprises the steps that character samples of a language to be studied are collected, and each character sample corresponds to a preset game operation instruction; then, a local or remote voice signal is received, and voice recognition is performed on the voice signal, and the voice signal isconverted into a character signal; the character signal and the character samples are cooperatively processed, and the game operation instruction corresponding to the matched character sample is called; according to the game operation instruction, the game operation is controlled; according to the game result, the studying result is evaluated; the man-machine conversation function is fused, and auser can perform voice answering according to voice problems or character problems in the same and emit voice signals for interaction; accordingly, the language studying is combined into the game, studying and entertainment are integrated, the user's studying initiative is greatly improved, and the studying effect is better.

The invention belongs to the technical field of voice recognition, and discloses a content recommendation method and system based on voice searching. The method comprises the following steps of obtaining voice problems continuously input by users; when the number of times of continuously inputting the same voice problem exceeds a preset threshold value, recognizing the identity of users continuously inputting the same voice problem, and determining the affiliated target subject of the voice problems; when the users continuously inputting the same voice problems are one user, inquiring the mastering degree of the user on the target subject according to the identity of the user; selecting the learning content with the difficulty grade matched with the mastering degree from the target subjectto be used as a target learning content; and outputting the target learning content. The learning content with the difficulty grade matched with the mastering degree is selected to be used as the target learning content to be recommended to the user, so that the learning interestingness reduction of students due to the output of the same answer is avoided; and meanwhile, a directing method can befavorably regulated according to the practical conditions of students so that the learning guidance effect of learning equipment can be achieved.

The invention provides a man-machine interaction logistics robot and a control method thereof. The method comprises the following steps: acquiring voiceprint information of a voice interaction command and a corresponding intention logistics task; calculating the matching degree between the voiceprint information of the voice interaction command and a plurality of pieces of preset voiceprint information, and if the voiceprint information of the voice interaction command is matched with one piece of preset voiceprint information, passing the identity verification of the user; matching the intention logistics task with tasks in a preset logistics task list, selecting the task with the highest matching degree as an optimal matching task, and sending a randomly generated verification code to a user mobile phone number recorded in the optimal matching task; and sending out a voice question according to the optimal matching task, receiving answer voice information of the user for the voice question, comparing the answer voice information with a starting point, a destination point and a randomly generated verification code recorded in the optimal matching task, and executing logistics transportation according to the optimal matching task if the answer voice information is matched with the starting point and the destination point recorded in the optimal matching task. According to the method and the device, safety control in man-machine interaction of the logistics robot is realized, and occurrence of error commands is prevented.

The invention discloses a voice-based dynamic verification method and software. When one system needs dynamic verification, the software embeds a voice question or a voice instruction into a verification operation page, a user plays voice, inputs a question answer or an instruction response into the verification operation interface according to the question or the instruction and submits the question answer or instruction response, and the software verifies whether the answer or instruction response is correct or not and returns a verification result to the system. According to the invention,dynamic verification is realized through voice, and the safety of the system is improved.

The invention discloses an online course content detection method and system, and the method comprises the steps that when a user side repeatedly plays the content of an online course video in a certain time period of an online course, a server collecting the playing frequency of the content and detects whether the content of the online course video in the time period has content required to be optimized or not according to a collection result. According to the invention, knowledge difficulties existing in the online course are screened out according to the frequency and the duration of watching a certain content of the online course video by a student at the user side, and whether the knowledge difficulties are online course voice explanation defects is determined by detecting the audio, so that the quality of the online course is improved, and the time waste caused by the voice problem of the student is shortened.

The invention discloses a visual impaired crowd-oriented scene character interactive understanding system, which comprises a mobile phone mobile terminal APP and a rear-end visual interactive calculation processing platform, the mobile phone mobile terminal APP comprises a voice wake-up component, a visual scene shooting component, a voice problem acquisition component, a voice-to-text conversion component, a logic judgment component, a data transmission transceiving component and a text-to-voice synthesis component; the back-end visual interaction processing computing platform comprises an input preprocessing module and a multi-head attention mechanism model. According to the method and the device, the character information in different scene pictures can be identified, the scene information is autonomously acquired by a user, the environmental adaptability and the expandability are high, and the sensitivity and the accuracy of character identification in the scene are high. Dynamic answering can be carried out according to user questions, and practicability and real-time performance are higher; the system can be installed on a mobile phone mobile terminal, can carry out information interaction through voice, and is convenient to use, low in cost consumption and easy to use for a user.

The invention discloses a voice help prompt method of a network community. The method comprises the following steps: providing a database, and storing a plurality of problems and corresponding solutions; unfolding a network community interface; in the interface, activating a voice help option; turning on a microphone to input a voice problem; matching the voice problem with all problems in the database; playing the problem with a highest matching degree by voice, and prompting a user to input the voice: answering yes or no about whether the problem is correct or not; if a user voice input result is yes, playing a solution corresponding to the problem with the highest matching degree by voice; and if the user voice input result is no, playing the voice problem which is input by the user again through voice, and returning to the above voice problem input step. Voice help is searched through voice input to quickly find the solution corresponding to the voice problem so as to bring convenience for users to use.

The invention relates to a method and a system for data processing based on a speech recognition technology. The method comprises the following steps: performing feature extraction on text informationto generate a plurality of feature words; determining the matching degree of the text information and the function description information based on the matching degree of each feature word and the function description information of the terminal equipment; when the matching degree of the text information and the function description information is greater than a matching degree threshold value, outputting a prompt tone for indicating to start data processing through a loudspeaker of the terminal equipment, and displaying a prompt message for indicating to start data processing through a display of the terminal equipment; acquiring a voice question input by a user through a voice recognition component of terminal equipment, converting the voice question into a text question, and performingretrieval in a local database based on the text question to determine reply content used for responding to the text question.

The invention discloses a voice and text transcriptional analysis-combined depression auxiliary diagnosis method and system and a storage medium. The method comprises the steps: carrying out the interaction with a target user through a voice question corresponding to a standardized question, and determining corresponding inquiry information; performing prediction through a preset algorithm based on the inquiry information; and judging whether the target user is a depression patient or not according to a prediction result. According to the invention, possible depression patients can be identified in an assisted manner, the workload of medical staff is reduced, and the depression diagnosis efficiency is improved.

The invention provides a computer motherboard which can be prompted by voice. The board of the computer motherboard is provided with a small voice device. When the computer is in trouble, the motherboard will automatically broadcast the voice problem, so that we can conveniently process and solve the problem.

The invention discloses a speech enhancement method, electronic equipment and a storage medium, and the speech enhancement method comprises the steps: converting a to-be-enhanced audio signal into a bitmap; and the bitmap is processed by using a neural network image processing technology, and the neural network image processing technology comprises a neural network based on image noise reduction. The to-be-enhanced audio signal is converted into the bitmap, and then the bitmap is processed by using the neural network image processing technology, so that the voice problem can be processed from the perspective of image processing, the research and development progress can be successfully accelerated by using deep learning in the field of image processing, and further, the development efficiency is improved. The neural network based on image noise reduction obtains a better speech enhancement effect.