128 results about "Single sentence" patented technology

A voice recognition method for a single sentence including a multi-instruction in an interactive voice user interface, method includes steps of detecting a connection ending by analyzing the morphemes of a single sentence on which voice recognition has been performed, separating the single sentence into a plurality of passages based on the connection ending, detecting a multi-connection ending by analyzing the connection ending and extracting instructions by specifically analyzing passages including the multi-connection ending and outputting a multi-instruction included in the single sentence by combining the instructions extracted in the step of extracting instructions. In accordance with the present invention, consumer usability can be significantly increased because a multi-operation intention can be checked in one sentence.

The invention discloses network criticism oriented viewpoint subject identifying method and system. The method comprises the following steps of: a. text inputting: inputting a criticism source and all criticism texts; b. text preprocessing: carrying out word division and speech part marking on the input texts, removing stop words, punctuations and specific empty words and calculating the word frequency information of the words; c. subject word judgment: calculating a word weight value, and if the word weight value is larger than a set threshold value, judging that a subject word is a viewpoint subject word; d. subject constructing: combining scattered viewpoint subject words into an integrated viewpoint subject; e. subject screening: confirming an effective viewpoint subject by filtering the viewpoint subject. The invention overcomes the field limitation of viewpoint analyzing method and system, identifies the viewpoint subject in a whole angle without constructing a body library, effectively overcomes the difficulty existing in single-sentence viewpoint analysis and automatically identifies the viewpoint subject in a phrase mode of a wide field and network criticism data which are dynamically changed along with the time.

The invention discloses a Chinese text automatic correction method. The method comprises the following steps of: a) inputting a to-be-corrected Chinese text, and performing word segmentation preprocessing on the Chinese text sentence by sentence; b) searching for one-character words, two-character words or disperse strings of three or more than three characters occurring in the text subjected to word segmentation sentence by sentence; c) performing continuous determination on the disperse strings occurring in the text subjected to word segmentation by adopting an N-gram model, and checking text word level errors for each single sentence in combination with a word forming probability of separate characters; and d) constructing an error correction knowledge base to generate an error correction candidate text. According to the Chinese text automatic correction method provided by the invention, the one-character words, two-character words or disperse strings of three or more than three characters occurring in the text subjected to word segmentation are searched for sentence by sentence, the disperse strings occurring in the text subjected to word segmentation are subjected to continuous determination by adopting the N-gram model to determine identification errors, and the error correction knowledge base is constructed to generate the error correction candidate text, so that error checking and correcting processes are combined very well, and the method has the characteristics of high error checking speed and high error correcting efficiency.

The invention comprises a bimodal man-man conversation sentiment analysis system and a bimodal man-man conversation sentiment analysis method based on machine learning. The bimodal man-man conversation sentiment analysis system is characterized by comprising a speech recognition module, a text deep-layer feature extraction module, a speech segmentation module, an acoustic feature extraction module, a feature fusion module and an sentiment analysis module, wherein the speech recognition module is used for recognizing speech content and a time label; the text deep-layer feature extraction module is used for completing the extraction of text deep-layer word level features and text deep-layer sentence level features; the speech segmentation module is used for segmenting single sentence speech from entire speech; the acoustic feature extraction module is used for completing the extraction of acoustic features of the speech; the feature fusion module is used for fusing the obtained text deep-layer features with the acoustic features; and the sentiment analysis module is used for acquiring sentiment polarities of the speech to be subjected to sentiment analysis. The bimodal man-man conversation sentiment analysis method can integrate the text and audio modals for recognizing conversation sentiment, and fully utilizes features of word vectors and sentence vectors, thereby improving the precision of recognition.

A method for displaying an internet web page with an area of focus to reduce viewer distraction. The selected area of focus is displayed with the normal text, graphic and background color attributes as created by the web page designer. The area of focus may be, for example, a single sentence or a paragraph and may include one or more frames from the web page. Shading is applied to other portions of the displayed web page which become a less pronounced area out of focus. The area of focus may be moved with a pointing device such as a mouse or a keyboard. The degree of shading may be set by the computer user. The computer displaying the web page may be programmed to read the text in the area of focus with synthesized speech and to move the area of focus as the text is read.

The invention discloses a real-time feedback method and a system of long voice continuous recognition and a recognition result. The method comprises the following steps: extension possibility from a sentence end point to an identification start point in an identification network is increased; in the process of decoding, when an identification path extends, the identification path can not only generate a single sentence, but also generate a plurality of continuous sentences, and large space video signals are separated through overall optimization of acoustics and linguistic probability; common parts of optimal history path of all active nodes are detected at fixed period, and identification word order which is fixed at the present moment is obtained; updated local identification result is fed back to a user in real time; and decoding space which is corresponding to identification parts and ensured is recycled. When the real-time feedback method and the system of the long voice continuous recognition and the recognition result are used, endpoint detection algorithm is not needed, continuous and sustainable identification of long voice signals are achieved, and human-computer interaction experience is obtained for users.

The embodiment of the invention provides a character relation identification method and device and a word segmentation method and relates to the technical field of internet information processing. The method comprises the steps that word segmentation processing is conducted on input texts to obtain word segmentation results; part-of-speech tagging is conducted on segmented words in the word segmentation results; corresponding grammatical components of the segmented words in single sentences are determined to generate a grammatical tree; the segmented words conforming to preset segmented word screening rules are extracted to generate an entity set; the entity set and the grammatical tree are compared and model simulation is performed to generate an entity relation diagram; a character relation diagram is obtained according to the entity relation diagram. Compared with the character relation diagram establishment process in the prior art, the method has the advantages of being small in operating amount, high in operating efficiency, short in consumed time and low in achievement difficulty.

The present invention provides a speech sound translating method and apparatus, according to one aspect of the invention, provided is a speech sound translating method, comprising: recognizing the speech sound as a text which has at least one long sentence comprising a plurality of single sentences; cutting at least one long sentence into a plurality of single sentences; translating each of a plurality of single sentences into a sentence of target language kind. The invention can insert a long sentence cutting module between a speech sound recognition module and a machine translation module, can recognize the long sentence in the text and can automatically cut the long sentences into a plurality of simple and complete single sentences, thereby reducing the difficulty of the machine translation and improving the quality of the machine translation. Moreover, the invention also provides an interface interactive with the user, enables the user to conveniently modify the automatic cutting result, meanwhile, the modifying operation of the user is recorded for updating the cutting model in real time, and the automatic cutting effect is improved continuously.

The invention discloses a visual document atlas construction method, which comprises the following steps of S1, sequentially extracting keywords from an input text, carrying out sentence segmentationon the input text, carrying out word segmentation on each single sentence generated by sentence segmentation, and sequentially carrying out part-of-speech tagging, named entity identification and dependency syntax analysis on each word formed by word segmentation; S2, formulating a relationship extraction rule, and extracting triple data from each simple sentence obtained in S1 based on the relationship extraction rule, wherein the triple data is composed of two entity words and a relation word; S3, importing the triple data obtained in the S2 into a graph database to form a document graph; and S4, performing graph mining operation on the data in the graph database, and realizing document graph visualization based on the mining operation. According to the visual document atlas constructionmethod, the key information of the text can be extracted, and the document is mapped into the visual graph based on semantic association, so that a user is helped to efficiently master the semantic information of the article.

The invention relates to a dialogue system-oriented data processing method and device. The data processing method includes: a first step, acquiring dialogue interaction information input by a user; a second step, generating a corresponding feature vector according to the dialogue interaction information and context interaction data, using the feature vector to search dialogue groups stored in a preset knowledge base, and acquiring feedback information corresponding to a matched dialogue group; and a third step, generating feedback information for the dialogue interaction information according to the feedback information corresponding to the matched dialogue group. The dialogue system-oriented data processing method can perform search in the knowledge base according to the content referred in a previous dialogue process of two dialogue parts, and is not like the conventional methods which perform search in the knowledge dependent on single sentences to generate the feedback information; and accordingly, compared with the conventional methods, the method and device can effectively avoid the problem that the obtained feedback information does not match the dialogue interaction information, and in this way, the user experience of the dialogue system can be improved.

The invention provides a multi-task chapter-level event extraction method based on a multi-headed self-attention mechanism. The method comprises the following steps: converting single sentence-level event extraction into chapter-level event extraction of a packaged sentence set; carrying out the word embedding representation by utilizing a pre-trained language model BERT; taking all word embedding and position embedding in a single sentence as input, employing a convolutional neural network model for coding, and capturing the most valuable features in the sentence in combination with a segmented maximum pool strategy; utilizing a multi-head self-attention model to obtain chapter representation and attention weight fused with full-text semantic information; utilizing a classifier to obtain a predicted event type; taking event types as prior information, linking the event types to an input sequence for event element extraction, and extracting all related elements in the sequence by using a pre-training model in combination with a machine reading understanding method. The method can be used for text-level event extraction tasks, and the breakthrough of converting a sequence labeling problem into a machine reading understanding problem is achieved.

The invention discloses a method and device for analyzing the semantic orientation of a Chinese network topic comment text. The method comprises the following steps: performing word segmentation and sentence segmentation on the Chinese network topic comment text to obtain a result sequence; performing syntactic analysis and grammatical analysis on the result sequence to obtain an evaluation object; performing sentence pattern analysis on the result sequence to determine simple sentences and complex sentences in the comment text, judging the relations among all the simple sentences forming a complex sentence, and determining a first emotion orientation value of sentence pattern analysis; extracting emotion phrases in each sentence in the result sequence according to the evaluation object and a preset phrase matching mode, and calculating a second emotion orientation value of each emotion phrase; calculating a third emotion orientation value of each sentence in the comment text according to the first emotion orientation value and the second emotion orientation values; determining a text emotion orientation value of the comment text according to the third emotion orientation values. According to the method, the accuracy and the recall rate of the semantic orientation analysis of the network topic comment text are improved.

The invention discloses a text abstract generation method based on a K-means model and a neural network model, and the method comprises the steps of preprocessing an original text, obtaining single sentences and words through segmentation, inputting the sentences and words into a doc2vec model, and carrying out the training, so as to obtain sentence vectors; determining the number of clustering centers of the original text, inputting the sentence vector into an unsupervised K-means model, and training to obtain a clustering center vector; calculating the Euclidean distance between the clustering center vector and the sentence vector, and extracting a sentence corresponding to the sentence vector closest to the clustering center to serve as a reference abstract; and inputting the original text, the reference abstract and the words into a generative neural network model to generate a text abstract. The method has the beneficial effects that the unsupervised model and the supervised neural network model are combined, so that the generated text abstract can be semantically coherent and is convenient for a user to understand.

The invention relates to a Chinese-Vietnamese unsupervised neural machine translation method fusing an EMD minimized bilingual dictionary, and belongs to the technical field of machine translation. The method comprises the steps of collecting corpora; crawling Chinese and Vietnamese single sentences by using a web crawler; firstly, training monolingual word embedding of Chinese and Vietnamese respectively, and obtaining a Chinese-Vietnamese bilingual dictionary through EMD training of minimized word embedding distribution; taking the dictionary as a seed dictionary for training to obtain Chinese-Vietnamese bilingual word embedding; and finally, embedding the bilingual words into an unsupervised machine translation model of a shared encoder to construct the Chinese-Vietnamese neural machinetranslation method fusing the EMD minimized bilingual dictionary. According to the method, the performance of the Hami unsupervised neural machine translation can be effectively improved.

The invention provides a Chinese proofreading and error-correction method and system based on Chinese word segmentation. The method comprises the steps of: carrying out Chinese word segmentation on input Chinese text according to single sentences to obtain word arrays, wherein the word arrays comprise single-character words, two-character words, three-character words or four-character words; recombining the word arrays to form short sentences; judging whether occurrence frequency of a short sentence in a preset text library is greater than a first threshold, and if yes, marking the short sentence as correct; and if not, carrying out glyph and Pinyin error-correction processing on the short sentence. The method realizes automatic proofreading and error correction of wrongly written characters in the text, and improves accuracy and work efficiency of Chinese proofreading and error correction.

For a given sentence grammar, speech recognizers are often required to decode M sets of HMMs each of which models a specific acoustic environment. In order to match input acoustic observations to each of the environments, typically recognition search methods require a network of M sub-networks. A new speech recognition search method is described here, which needs a network that is only the size of a single sub-network and yet provides the same recognition performance, thus reducing the memory requirements for network storage by (M-1) / M.

The invention discloses a method and a device for extracting the subject term from a simple sentence, which belongs to the technical field of subject term extraction. The method comprises the following steps: counting a plurality of key words and a plurality of multielement combinations in a language material, working out the characteristic value of each key word, and determining the sequence of each multielement combination; taking each key word of a simple sentence as the current key word respectively, extracting the multielement combination containing the current key word, calculating the grade of the current key word according to the sequence of the extracted multielement combination, and working out the metric of the current key word according to the characteristic value and the grade of the current key word; and picking out the subject term of the simple sentence from all key words according to the metrics after the metrics of all the key words of the single sentence are obtained. The device comprises a counting module, a characteristic value calculation module, a sequence calculation module, an extraction module, a grade calculation module, a metric calculation module and a subject term selection module. The invention has the advantage that the text message of a simple sentence is fully utilized, so that the accuracy and the efficiency of extracting the subject term from a simple sentence can be improved.

The invention relates to the technical field of artificial intelligence, in particular to a text review method and device, electronic equipment and a computer storage medium. The method comprises thesteps of obtaining an examination document, and a document type and a document format of the examination document; calling a system database, wherein key auditing terms are stored in the system database; performing clause processing on the examination document to form a plurality of simple sentences; inputting each simple sentence into a system database for matching to obtain sentence similarity of each simple sentence; judging whether the similarity of each sentence is greater than a set threshold value or not, and marking the single sentence when the similarity is greater than the set threshold value. According to the method, the corresponding system database is correspondingly called through the document type and the document format, statement similarity calculation is carried out on the simple sentence of the to-be-examined text and the key examination terms in the system database, the corresponding statement is labeled according to the similarity, risk prompt information can be carried out, the text examination risk is reduced, and the text examination efficiency is improved.

The invention discloses a conversation management method, device and system based on verbal skill strategy management and a computer readable medium, and the method comprises the steps: enabling an intelligent voice robot to select a theme to call a user, enabling the user to respond to the call of the intelligent voice robot, judging the conversation state of the user and the intelligent voice robot, if the call state is user response, identifying that the user voice is converted into a user voice text, determining the real intention of the user according to the user voice text, and selectinga matched dialogue strategy according to the determined real intention of the user, selecting corresponding dialogue content based on the dialogue strategy, and outputting the dialogue content to theuser. By adopting the technical scheme, the intelligent voice robot not only can identify the simple sentence intention expressed by the current sentence of the user, but also can understand the realintention of the user based on the previous sentence of the user, so that the user experience is better.

The invention provides an intelligent lyric generation method. The intelligent lyric generation method comprises the steps of obtaining a preset single sentence of lyric; obtaining a keyword list of the to-be-generated lyric set; inputting the single sentence of lyrics and the keyword list into a pre-trained lyric generation model to generate a next sentence of lyrics; and taking the generated next sentence of lyric as the preset single sentence of lyric, and repeatedly executing the step of inputting the single sentence of lyric and the keyword list into a pre-trained lyric generation model to generate the next sentence of lyric so as to generate a lyric set. According to the method, brand-new and non-existing lyrics can be intelligently generated without preparing a lyric corpus in advance, and lyric creation is achieved.

An information analysis apparatus, an information analysis method, and a program are provided that enable target information to be determined in units of single sentences, rather than in units of plural sentences, while taking into consideration the tendency of appearance of the target information. An information analysis apparatus 5 is used to perform an analysis on text information to determine whether or not the text information corresponds to the target information. The information analysis apparatus 5 includes a density estimation unit 51 that estimates, in units of analysis each composed of a plurality of sentences of text information, a density indicating the degree to which the target information is included in the unit of analysis, and a determination unit 52 that obtains an evaluation value indicating the degree to which each sentence included in each unit of analysis corresponds to the target information from the estimated density of the unit of analysis, and determines whether or not the sentence corresponds to the target information based on the evaluation value.

The invention discloses a word-sentence-level short text classification method based on deep learning, and belongs to the technical field of natural language processing. According to the method, wordfeatures and sentence features are combined based on a word vector technology to express complex text features. By performing convolution pooling on the word vectors in the single sentence through a plurality of convolution kernels of the convolutional neural network, all the feature maps are connected to obtain sentence vectors, and the sentence vectors are inputted into the long-term and short-term memory network according to a time sequence to perform context association so as to better express text contents. To-be-classified short text data is subjected to sentence segmentation, word segmentation, stop word removal, word vector conversion and the like and then input into a word-sentence level convolutional recurrent neural network for training, and finally, a short text classificationmodel can be obtained and a short text classification task is completed. The method has good performance in the aspects of tested Chinese spam e-mail classification and news text classification.

A rapid translation method comprises the step of allocating a file to be translated to single sentences to be translated in advance, and further comprises the step of comparing the single sentences to be translated with single sentences in a historical memory vault, the first screening step of judging whether the single sentences to be translated appear in the file to be translated for the first time or not, and the recovering step of recovering a translation of all the single sentences in the historical memory vault and a first screening library according to the corresponding relation to be the corresponding positions of the file to be translated after translating is completed. A rapid translation device comprises the historical memory vault, an input allocation module, a first judgment module, a comparison module, the first screening library, an identification module, a recovery module and a display device. The comparison module is connected with the historical memory vault and compares the single sentences in the vault with each single sentence to be translated obtained by allocation of the input allocation module. The first judgment module judges each single sentence obtained by allocation of the input allocation module. According to the rapid translation device and method, the number of characters to be translated is reduced, and the quality of the translation is ensured.

The invention provides an interaction method and system based on an intelligent robot. The method comprises the following steps that dialogue data from a user is received and is parsed, and whether preset expression information in the dialogue data is defaulted or not is monitored; under the situation that the preset expression information in the dialogue data is defaulted, and a defaulted dialogue mechanism is invoked to generate dialogue data. The method further comprises the following steps that a dialogue data structure is configured, and configuration parameters are utilized to generate dialogue data under the data structure by combining with context configuration parameters. Multiple cycles of interaction are performed by using the robot and invoking the defaulted dialogue mechanism, all required information parameters can be obtained through multiple cycles of dialogues under the situation that single-sentence expression contents of the user do not provide all the information parameters, reasonable and accurate information can be further provided to the user, user experience can be improved, and the interaction demand of the user is met.

A voice data creation system includes a dictionary data memory section that stores dictionary data for generating synthesized voice data corresponding to text data; an edition processing section that displays an edition screen for editing a voice guidance message as a sentence including a plurality of phrases to receive edition input formation so as to perform an edition processing based on the edition input information; a list information generation processing section that generates list information relating to each sentence and phrases included in the each sentence based on a result of the edition processing; a phrase voice data generating section that determines a target phrase for voice data creation based on the list information to generate and maintain voice data corresponding to the target phrase determined for voice data creation based on the dictionary data; and a memory write information generating section that determines a target phrase to be stored in a voice data memory based on the list information to generate memory write information including voice data of the target phrase determined to be stored in the memory. In the system, the edition processing section divides a sentence input on the edition screen into a plurality of phrases based on text data of the sentence input; the list information generation processing section specifies the phrases included in the sentence and a reproduction order of the phrases based on a result of the sentence division to generate sentence information including phrase specification information of the phrases included in the sentence and sequence information relating to the reproduction order of the phrases; the phrase voice data generating section generates synthesized voice data corresponding to text data of the target phrase for voice data creation based on the dictionary data; and the memory write information generating section determines the target phrase to be stored in the memory such that voice data of a phrase used commonly in a plurality of sentences and a phrase used a plurality of number of times in a single sentence are not duplicately stored.

The present invention relates to a computer assisted foreign language reading training system and method. The present system includes: a reading operation window which can display at least one foreign language passage for learners to practice reading, a vocabulary classification unit which can display words in the passage in special display effects according to classifications determined by learners, a single sentence translation unit which translates the sentences in the passage and displays the corresponding Chinese translation, a sentence reorganizing unit which can change sentence orders for learners to practice reorganizing sentences, and a database which provides the reading operation window, the vocabulary classification unit, the single sentence translation unit and the sentence reorganizing unit with data or files to enhance the vocabulary ability and logic judgment ability of learners so as to improve their foreign language reading ability. The present method helps English learners to improve their reading skills through first reading, intensive reading, understanding, retelling as well as assessing.

The invention discloses a generation method and device of a synonymous sentence generation model, and a medium. The method comprises the steps of performing training by using a first set to obtain a generation model and a discrimination model; wherein the first set comprises a plurality of synonymous sentence groups, and each synonymous sentence group comprises at least two synonymous sentences; and performing iterative processing on the generation model and the discrimination model until the generation model converges. A large number of simple sentences are used, and the advantages of low cost and no need of manual annotation of the simple sentences are fully utilized. In the training process of the model, a large number of simple sentences and a reinforcement learning mode are combined,the semantic richness is greatly improved through the use of the simple sentences, the reinforcement learning mode can enable the model to be continuously optimized in the iteration process, and therefore a high-quality synonymous sentence generation model can be trained completely without depending on a large number of parallel corpora.

The invention discloses a difference item discrimination method and device, and the method comprises the steps: obtaining a target difference item in a recognition result of a first simple sentence and a recognition result of a second simple sentence, wherein the target difference item comprises a first difference text and a second difference text, the first simple sentence comprises a public itemand a first difference text, and the second simple sentence comprises a public item and a second difference text; determining a first probability corresponding to the first difference text and a second probability corresponding to the second difference text based on the language prediction model and the common item; and judging whether the target difference item is a real difference item or not according to the first probability and the second probability. By implementing the method and the device, the real difference item can be effectively judged, and the non-real difference item caused bythe OCR identification error is filtered, so that the accuracy of single sentence comparison is improved.

The invention discloses a sentence distance mapping method and device based on machine learning, computer equipment and a storage medium. The method comprises the steps that input single sentence voice information is acquired; converting the single-sentence voice information into single-sentence text information; preprocessing the single sentence text information, and querying a preset word vectorlibrary to obtain word vectors corresponding to words in the preprocessed single sentence text information; calculating the distance between the single sentence text information and a preset standardsingle sentence by using a preset algorithm according to the word vector corresponding to each word in the single sentence text information; and inputting the distance into a preset function, mappinga score, and training the preset function through training data. Therefore, the similarity between the sentences is accurately calculated, and the technical effect of being more accurate and more visual is achieved.