Generating Factual Text via Entailment Recognition Task

Generating diverse and factual text is challenging and is receiving increasing attention.By sampling from the latent space,variational autoencoder-based models have recently enhanced the diversity of generated text.However,existing research predominantly depends on summarizationmodels to offer paragraph-level semantic information for enhancing factual correctness.The challenge lies in effectively generating factual text using sentence-level variational autoencoder-based models.In this paper,a novel model called fact-aware conditional variational autoencoder is proposed to balance the factual correctness and diversity of generated text.Specifically,our model encodes the input sentences and uses them as facts to build a conditional variational autoencoder network.By training a conditional variational autoencoder network,the model is enabled to generate text based on input facts.Building upon this foundation,the input text is passed to the discriminator along with the generated text.By employing adversarial training,the model is encouraged to generate text that is indistinguishable to the discriminator,thereby enhancing the quality of the generated text.To further improve the factual correctness,inspired by the natural language inference system,the entailment recognition task is introduced to be trained together with the discriminator via multi-task learning.Moreover,based on the entailment recognition results,a penalty term is further proposed to reconstruct the loss of our model,forcing the generator to generate text consistent with the facts.Experimental results demonstrate that compared with competitivemodels,ourmodel has achieved substantial improvements in both the quality and factual correctness of the text,despite only sacrificing a small amount of diversity.Furthermore,when considering a comprehensive evaluation of diversity and quality metrics,our model has also demonstrated the best performance.

E2E-MFERC:AMulti-Face Expression Recognition Model for Group Emotion Assessment

In smart classrooms, conducting multi-face expression recognition based on existing hardware devices to assessstudents’ group emotions can provide educators with more comprehensive and intuitive classroom effect analysis,thereby continuouslypromotingthe improvementof teaching quality.However,most existingmulti-face expressionrecognition methods adopt a multi-stage approach, with an overall complex process, poor real-time performance,and insufficient generalization ability. In addition, the existing facial expression datasets are mostly single faceimages, which are of low quality and lack specificity, also restricting the development of this research. This paperaims to propose an end-to-end high-performance multi-face expression recognition algorithm model suitable forsmart classrooms, construct a high-quality multi-face expression dataset to support algorithm research, and applythe model to group emotion assessment to expand its application value. To this end, we propose an end-to-endmulti-face expression recognition algorithm model for smart classrooms (E2E-MFERC). In order to provide highqualityand highly targeted data support for model research, we constructed a multi-face expression dataset inreal classrooms (MFED), containing 2,385 images and a total of 18,712 expression labels, collected from smartclassrooms. In constructing E2E-MFERC, by introducing Re-parameterization visual geometry group (RepVGG)block and symmetric positive definite convolution (SPD-Conv) modules to enhance representational capability;combined with the cross stage partial network fusion module optimized by attention mechanism (C2f_Attention),it strengthens the ability to extract key information;adopts asymptotic feature pyramid network (AFPN) featurefusion tailored to classroomscenes and optimizes the head prediction output size;achieves high-performance endto-end multi-face expression detection. Finally, we apply the model to smart classroom group emotion assessmentand provide design references for classroom effect analysis evaluation metrics. Experiments based on MFED showthat the mAP and F1-score of E2E-MFERC on classroom evaluation data reach 83.6% and 0.77, respectively,improving the mAP of same-scale You Only Look Once version 5 (YOLOv5) and You Only Look Once version8 (YOLOv8) by 6.8% and 2.5%, respectively, and the F1-score by 0.06 and 0.04, respectively. E2E-MFERC modelhas obvious advantages in both detection speed and accuracy, which can meet the practical needs of real-timemulti-face expression analysis in classrooms, and serve the application of teaching effect assessment very well.

Named Entity Recognition for Nepali Text Using Support Vector Machines

Named Entity Recognition aims to identify and to classify rigid designators in text such as proper names, biological species, and temporal expressions into some predefined categories. There has been growing interest in this field of research since the early 1990s. Named Entity Recognition has a vital role in different fields of natural language processing such as Machine Translation, Information Extraction, Question Answering System and various other fields. In this paper, Named Entity Recognition for Nepali text, based on the Support Vector Machine (SVM) is presented which is one of machine learning approaches for the classification task. A set of features are extracted from training data set. Accuracy and efficiency of SVM classifier are analyzed in three different sizes of training data set. Recognition systems are tested with ten datasets for Nepali text. The strength of this work is the efficient feature extraction and the comprehensive recognition techniques. The Support Vector Machine based Named Entity Recognition is limited to use a certain set of features and it uses a small dictionary which affects its performance. The learning performance of recognition system is observed. It is found that system can learn well from the small set of training data and increase the rate of learning on the increment of training size.

Enhanced Attention-Based Encoder-Decoder Framework for Text Recognition

Recognizing irregular text in natural images is a challenging task in computer vision.The existing approaches still face difficulties in recognizing irre-gular text because of its diverse shapes.In this paper,we propose a simple yet powerful irregular text recognition framework based on an encoder-decoder archi-tecture.The proposed framework is divided into four main modules.Firstly,in the image transformation module,a Thin Plate Spline(TPS)transformation is employed to transform the irregular text image into a readable text image.Sec-ondly,we propose a novel Spatial Attention Module(SAM)to compel the model to concentrate on text regions and obtain enriched feature maps.Thirdly,a deep bi-directional long short-term memory(Bi-LSTM)network is used to make a con-textual feature map out of a visual feature map generated from a Convolutional Neural Network(CNN).Finally,we propose a Dual Step Attention Mechanism(DSAM)integrated with the Connectionist Temporal Classification(CTC)-Attention decoder to re-weights visual features and focus on the intra-sequence relationships to generate a more accurate character sequence.The effectiveness of our proposed framework is verified through extensive experiments on various benchmarks datasets,such as SVT,ICDAR,CUTE80,and IIIT5k.The perfor-mance of the proposed text recognition framework is analyzed with the accuracy metric.Demonstrate that our proposed method outperforms the existing approaches on both regular and irregular text.Additionally,the robustness of our approach is evaluated using the grocery datasets,such as GroZi-120,Web-Market,SKU-110K,and Freiburg Groceries datasets that contain complex text images.Still,our framework produces superior performance on grocery datasets.

Region-Aware Fashion Contrastive Learning for Unified Attribute Recognition and Composed Retrieval

Clothing attribute recognition has become an essential technology,which enables users to automatically identify the characteristics of clothes and search for clothing images with similar attributes.However,existing methods cannot recognize newly added attributes and may fail to capture region-level visual features.To address the aforementioned issues,a region-aware fashion contrastive language-image pre-training(RaF-CLIP)model was proposed.This model aligned cropped and segmented images with category and multiple fine-grained attribute texts,achieving the matching of fashion region and corresponding texts through contrastive learning.Clothing retrieval found suitable clothing based on the user-specified clothing categories and attributes,and to further improve the accuracy of retrieval,an attribute-guided composed network(AGCN)as an additional component on RaF-CLIP was introduced,specifically designed for composed image retrieval.This task aimed to modify the reference image based on textual expressions to retrieve the expected target.By adopting a transformer-based bidirectional attention and gating mechanism,it realized the fusion and selection of image features and attribute text features.Experimental results show that the proposed model achieves a mean precision of 0.6633 for attribute recognition tasks and a recall@10(recall@k is defined as the percentage of correct samples appearing in the top k retrieval results)of 39.18 for composed image retrieval task,satisfying user needs for freely searching for clothing through images and texts.

Improving CNN-BGRU Hybrid Network for Arabic Handwritten Text Recognition

Handwriting recognition is a challenge that interests many researchers around the world.As an exception,handwritten Arabic script has many objectives that remain to be overcome,given its complex form,their number of forms which exceeds 100 and its cursive nature.Over the past few years,good results have been obtained,but with a high cost of memory and execution time.In this paper we propose to improve the capacity of bidirectional gated recurrent unit(BGRU)to recognize Arabic text.The advantages of using BGRUs is the execution time compared to other methods that can have a high success rate but expensive in terms of time andmemory.To test the recognition capacity of BGRU,the proposed architecture is composed by 6 convolutional neural network(CNN)blocks for feature extraction and 1 BGRU+2 dense layers for learning and test.The experiment is carried out on the entire database of institut für nachrichtentechnik/ecole nationale d’ingénieurs de Tunis(IFN/ENIT)without any preprocessing or data selection.The obtained results show the ability of BGRUs to recognize handwritten Arabic script.

Mathematical Named Entity Recognition Based on Adversarial Training and Self-Attention

Mathematical named entity recognition(MNER)is one of the fundamental tasks in the analysis of mathematical texts.To solve the existing problems of the current neural network that has local instability,fuzzy entity boundary,and long-distance dependence between entities in Chinese mathematical entity recognition task,we propose a series of optimization processing methods and constructed an Adversarial Training and Bidirectional long shortterm memory-Selfattention Conditional random field(AT-BSAC)model.In our model,the mathematical text was vectorized by the word embedding technique,and small perturbations were added to the word vector to generate adversarial samples,while local features were extracted by Bi-directional Long Short-Term Memory(BiLSTM).The self-attentive mechanism was incorporated to extract more dependent features between entities.The experimental results demonstrated that the AT-BSAC model achieved a precision(P)of 93.88%,a recall(R)of 93.84%,and an F1-score of 93.74%,respectively,which is 8.73%higher than the F1-score of the previous Bi-directional Long Short-Term Memory Conditional Random Field(BiLSTM-CRF)model.The effectiveness of the proposed model in mathematical named entity recognition.

Tibetan Multi-Dialect Speech Recognition Using Latent Regression Bayesian Network and End-To-End Mode

We proposed a method using latent regression Bayesian network (LRBN) toextract the shared speech feature for the input of end-to-end speech recognition model.The structure of LRBN is compact and its parameter learning is fast. Compared withConvolutional Neural Network, it has a simpler and understood structure and lessparameters to learn. Experimental results show that the advantage of hybridLRBN/Bidirectional Long Short-Term Memory-Connectionist Temporal Classificationarchitecture for Tibetan multi-dialect speech recognition, and demonstrate the LRBN ishelpful to differentiate among multiple language speech sets.

Digit Recognition in Natural Scene Texts

Digit recognition from a natural scene text in video surveillance/broadcasting applications is a challenging research task due to blurred, font variations, twisted, and non-uniform color distribution issues with a digit in a natural scene to be recognized. In this paper, to solve the digit number recognition problem, a principal-axis based topology contour descriptor with support vector machine （SVM） classification is proposed. The contributions of this paper include： a） a local descriptor with SVM classification for digit recognition, b） higher accuracy than the state-of-the art methods, and c） low computational power （0.03 second/digit recognition）, which make this method adoptable to real-time applications.

CNN and Fuzzy Rules Based Text Detection and Recognition from Natural Scenes

In today’s real world, an important research part in image processing isscene text detection and recognition. Scene text can be in different languages,fonts, sizes, colours, orientations and structures. Moreover, the aspect ratios andlayouts of a scene text may differ significantly. All these variations appear assignificant challenges for the detection and recognition algorithms that are consideredfor the text in natural scenes. In this paper, a new intelligent text detection andrecognition method for detectingthe text from natural scenes and forrecognizingthe text by applying the newly proposed Conditional Random Field-based fuzzyrules incorporated Convolutional Neural Network (CR-CNN) has been proposed.Moreover, we have recommended a new text detection method for detecting theexact text from the input natural scene images. For enhancing the presentation ofthe edge detection process, image pre-processing activities such as edge detectionand color modeling have beenapplied in this work. In addition, we have generatednew fuzzy rules for making effective decisions on the processes of text detectionand recognition. The experiments have been directedusing the standard benchmark datasets such as the ICDAR 2003, the ICDAR 2011, the ICDAR2005 and the SVT and have achieved better detection accuracy intext detectionand recognition. By using these three datasets, five different experiments havebeen conducted for evaluating the proposed model. And also, we have comparedthe proposed system with the other classifiers such as the SVM, the MLP and theCNN. In these comparisons, the proposed model has achieved better classificationaccuracywhen compared with the other existing works.

An Efficient Text Recognition System from Complex Color Image for Helping the Visually Impaired Persons

The challenge faced by the visually impaired persons in their day-today lives is to interpret text from documents.In this context,to help these people,the objective of this work is to develop an efficient text recognition system that allows the isolation,the extraction,and the recognition of text in the case of documents having a textured background,a degraded aspect of colors,and of poor quality,and to synthesize it into speech.This system basically consists of three algorithms:a text localization and detection algorithm based on mathematical morphology method(MMM);a text extraction algorithm based on the gamma correction method(GCM);and an optical character recognition(OCR)algorithm for text recognition.A detailed complexity study of the different blocks of this text recognition system has been realized.Following this study,an acceleration of the GCM algorithm(AGCM)is proposed.The AGCM algorithm has reduced the complexity in the text recognition system by 70%and kept the same quality of text recognition as that of the original method.To assist visually impaired persons,a graphical interface of the entire text recognition chain has been developed,allowing the capture of images from a camera,rapid and intuitive visualization of the recognized text from this image,and text-to-speech synthesis.Our text recognition system provides an improvement of 6.8%for the recognition rate and 7.6%for the F-measure relative to GCM and AGCM algorithms.

Embedded System Based Raspberry Pi 4 for Text Detection and Recognition

Detecting and recognizing text from natural scene images presents a challenge because the image quality depends on the conditions in which the image is captured,such as viewing angles,blurring,sensor noise,etc.However,in this paper,a prototype for text detection and recognition from natural scene images is proposed.This prototype is based on the Raspberry Pi 4 and the Universal Serial Bus(USB)camera and embedded our text detection and recognition model,which was developed using the Python language.Our model is based on the deep learning text detector model through the Efficient and Accurate Scene Text Detec-tor(EAST)model for text localization and detection and the Tesseract-OCR,which is used as an Optical Character Recognition(OCR)engine for text recog-nition.Our prototype is controlled by the Virtual Network Computing(VNC)tool through a computer via a wireless connection.The experiment results show that the recognition rate for the captured image through the camera by our prototype can reach 99.75%with low computational complexity.Furthermore,our proto-type is more performant than the Tesseract software in terms of the recognition rate.Besides,it provides the same performance in terms of the recognition rate with a huge decrease in the execution time by an average of 89%compared to the EasyOCR software on the Raspberry Pi 4 board.

An End-to-End Transformer-Based Automatic Speech Recognition for Qur’an Reciters

The attention-based encoder-decoder technique,known as the trans-former,is used to enhance the performance of end-to-end automatic speech recognition(ASR).This research focuses on applying ASR end-toend transformer-based models for the Arabic language,as the researchers’community pays little attention to it.The Muslims Holy Qur’an book is written using Arabic diacritized text.In this paper,an end-to-end transformer model to building a robust Qur’an vs.recognition is proposed.The acoustic model was built using the transformer-based model as deep learning by the PyTorch framework.A multi-head attention mechanism is utilized to represent the encoder and decoder in the acoustic model.AMel filter bank is used for feature extraction.To build a language model(LM),the Recurrent Neural Network(RNN)and Long short-term memory(LSTM)were used to train an n-gram word-based LM.As a part of this research,a new dataset of Qur’an verses and their associated transcripts were collected and processed for training and evaluating the proposed model,consisting of 10 h of.wav recitations performed by 60 reciters.The experimental results showed that the proposed end-to-end transformer-based model achieved a significant low character error rate(CER)of 1.98%and a word error rate(WER)of 6.16%.We have achieved state-of-the-art end-to-end transformer-based recognition for Qur’an reciters.

Text Augmentation-Based Model for Emotion Recognition Using Transformers

Emotion Recognition in Conversations(ERC)is fundamental in creating emotionally intelligentmachines.Graph-BasedNetwork(GBN)models have gained popularity in detecting conversational contexts for ERC tasks.However,their limited ability to collect and acquire contextual information hinders their effectiveness.We propose a Text Augmentation-based computational model for recognizing emotions using transformers(TA-MERT)to address this.The proposed model uses the Multimodal Emotion Lines Dataset(MELD),which ensures a balanced representation for recognizing human emotions.Themodel used text augmentation techniques to producemore training data,improving the proposed model’s accuracy.Transformer encoders train the deep neural network(DNN)model,especially Bidirectional Encoder(BE)representations that capture both forward and backward contextual information.This integration improves the accuracy and robustness of the proposed model.Furthermore,we present a method for balancing the training dataset by creating enhanced samples from the original dataset.By balancing the dataset across all emotion categories,we can lessen the adverse effects of data imbalance on the accuracy of the proposed model.Experimental results on the MELD dataset show that TA-MERT outperforms earlier methods,achieving a weighted F1 score of 62.60%and an accuracy of 64.36%.Overall,the proposed TA-MERT model solves the GBN models’weaknesses in obtaining contextual data for ERC.TA-MERT model recognizes human emotions more accurately by employing text augmentation and transformer-based encoding.The balanced dataset and the additional training samples also enhance its resilience.These findings highlight the significance of transformer-based approaches for special emotion recognition in conversations.

An Efficient Hybrid Model for Arabic Text Recognition

In recent years,Deep Learning models have become indispensable in several fields such as computer vision,automatic object recognition,and automatic natural language processing.The implementation of a robust and efficient handwritten text recognition system remains a challenge for the research community in this field,especially for the Arabic language,which,compared to other languages,has a dearth of published works.In this work,we presented an efficient and new system for offline Arabic handwritten text recognition.Our new approach is based on the combination of a Convolutional Neural Network(CNN)and a Bidirectional Long-Term Memory(BLSTM)followed by a Connectionist Temporal Classification layer(CTC).Moreover,during the training phase of the model,we introduce an algorithm of data augmentation to increase the quality of data.Our proposed approach can recognize Arabic handwritten texts without the need to segment the characters,thus overcoming several problems related to this point.To train and test(evaluate)our approach,we used two Arabic handwritten text recognition databases,which are IFN/ENIT and KHATT.The Experimental results show that our new approach,compared to other methods in the literature,gives better results.

Menu Text Recognition of Few-shot Learning

Recent advances in OCR show that end-to-end(E2E)training pipelines including detection and identification can achieve the best results.However,many existing methods usually focus on case insensitive English characters.In this paper,we apply an E2E approach,the multiplex multilingual mask TextSpotter,which performs script recognition at the word level and uses different recognition headers to process different scripts while maintaining uniform loss,thus optimizing script recognition and multiple recognition headers simultaneously.Experiments show that this method is superior to the single-head model with similar number of parameters in endto-end identification tasks.

An improved CRNN for Vietnamese Identity Card Information Recognition

This paper proposes an enhancement of an automatic text recognition system for extracting information from the front side of the Vietnamese citizen identity(CID)card.First,we apply Mask-RCNN to segment and align the CID card from the background.Next,we present two approaches to detect the CID card’s text lines using traditional image processing techniques compared to the EAST detector.Finally,we introduce a new end-to-end Convolutional Recurrent Neural Network(CRNN)model based on a combination of Connectionist Temporal Classification(CTC)and attention mechanism for Vietnamese text recognition by jointly train the CTC and attention objective functions together.The length of the CTC’s output label sequence is applied to the attention-based decoder prediction to make the final label sequence.This process helps to decrease irregular alignments and speed up the label sequence estimation during training and inference,instead of only relying on a data-driven attention-based encoder-decoder to estimate the label sequence in long sentences.We may directly learn the proposed model from a sequence of words without detailed annotations.We evaluate the proposed system using a real collected Vietnamese CID card dataset and find that our method provides a 4.28%in WER and outperforms the common techniques.

Speech Recognition via CTC-CNN Model

In the speech recognition system,the acoustic model is an important underlying model,and its accuracy directly affects the performance of the entire system.This paper introduces the construction and training process of the acoustic model in detail and studies the Connectionist temporal classification(CTC)algorithm,which plays an important role in the end-to-end framework,established a convolutional neural network(CNN)combined with an acoustic model of Connectionist temporal classification to improve the accuracy of speech recognition.This study uses a sound sensor,ReSpeakerMic Array v2.0.1,to convert the collected speech signals into text or corresponding speech signals to improve communication and reduce noise and hardware interference.The baseline acousticmodel in this study faces challenges such as long training time,high error rate,and a certain degree of overfitting.The model is trained through continuous design and improvement of the relevant parameters of the acousticmodel,and finally the performance is selected according to the evaluation index.Excellentmodel,which reduces the error rate to about 18%,thus improving the accuracy rate.Finally,comparative verificationwas carried out from the selection of acoustic feature parameters,the selection of modeling units,and the speaker’s speech rate,which further verified the excellent performance of the CTCCNN_5+BN+Residual model structure.In terms of experiments,to train and verify the CTC-CNN baseline acoustic model,this study uses THCHS-30 and ST-CMDS speech data sets as training data sets,and after 54 epochs of training,the word error rate of the acoustic model training set is 31%,the word error rate of the test set is stable at about 43%.This experiment also considers the surrounding environmental noise.Under the noise level of 80∼90 dB,the accuracy rate is 88.18%,which is the worst performance among all levels.In contrast,at 40–60 dB,the accuracy was as high as 97.33%due to less noise pollution.

An Auto-Grading Oriented Approach for Off-Line Handwritten Organic Cyclic Compound Structure Formulas Recognition

Auto-grading,as an instruction tool,could reduce teachers’workload,provide students with instant feedback and support highly personalized learning.Therefore,this topic attracts considerable attentions from researchers recently.To realize the automatic grading of handwritten chemistry assignments,the problem of chemical notations recognition should be solved first.The recent handwritten chemical notations recognition solutions belonging to the end-to-end trainable category suffered fromthe problem of lacking the accurate alignment information between the input and output.They serve the aim of reading notations into electrical devices to better prepare relevant edocuments instead of auto-grading handwritten assignments.To tackle this limitation to enable the auto-grading of handwritten chemistry assignments at a fine-grained level.In this work,we propose a component-detectionbased approach for recognizing off-line handwritten Organic Cyclic Compound Structure Formulas(OCCSFs).Specifically,we define different components of OCCSFs as objects(including graphical objects and text objects),and adopt the deep learning detector to detect them.Then,regarding the detected text objects,we introduce an improved attention-based encoder-decoder model for text recognition.Finally,with these detection results and the geometric relationships of detected objects,this article designs a holistic algorithm for interpreting the spatial structure of handwritten OCCSFs.The proposedmethod is evaluated on a self-collected data set consisting of 3000 samples and achieves promising results.

Recognition of Urdu Handwritten Alphabet Using Convolutional Neural Network (CNN)

Handwritten character recognition systems are used in every field of life nowadays,including shopping malls,banks,educational institutes,etc.Urdu is the national language of Pakistan,and it is the fourth spoken language in the world.However,it is still challenging to recognize Urdu handwritten characters owing to their cursive nature.Our paper presents a Convolutional Neural Networks(CNN)model to recognize Urdu handwritten alphabet recognition(UHAR)offline and online characters.Our research contributes an Urdu handwritten dataset(aka UHDS)to empower future works in this field.For offline systems,optical readers are used for extracting the alphabets,while diagonal-based extraction methods are implemented in online systems.Moreover,our research tackled the issue concerning the lack of comprehensive and standard Urdu alphabet datasets to empower research activities in the area of Urdu text recognition.To this end,we collected 1000 handwritten samples for each alphabet and a total of 38000 samples from 12 to 25 age groups to train our CNN model using online and offline mediums.Subsequently,we carried out detailed experiments for character recognition,as detailed in the results.The proposed CNN model outperformed as compared to previously published approaches.