Disfluency detection and reconstruction of spontaneous speech transcripts

The rapid spread of fake news has become a critical issue in the digital era, posing challenges to information authenticity and decision-making. This research presents a novel method for fake news detection leveraging a Generative Adversarial Network (GAN) combined with the BLEU (Bilingual Evaluation Understudy) score for evaluating textual quality. The proposed model uses a GAN framework to generate synthetic news data and trains a classifier to distinguish between genuine and fabricated articles. The BLEU score, commonly used in machine translation, is adapted to assess the accuracy of generated text against real-world news. Experimental results show the effectiveness of this approach, with detection performance categorized based on BLEU score ranges: Excellent (0.7+), Good (0.5-0.7), Average (0.3-0.5), and Poor (<0.3). With a BLEU score of 0.66, the model demonstrates strong performance in distinguishing fake from real news, with potential applications in automated content moderation and misinformation detection.

BackgroundLanguage barriers contribute significantly to health care disparities in the United States, where a sizable proportion of patients are exclusively Spanish speakers. In orthopedic surgery, such barriers impact both patients’ comprehension of and patients’ engagement with available resources. Studies have explored the utility of large language models (LLMs) for medical translation but have yet to robustly evaluate artificial intelligence (AI)–driven translation and simplification of orthopedic materials for Spanish speakers.ObjectiveThis study used the bilingual evaluation understudy (BLEU) method to assess translation quality and investigated the ability of AI to simplify patient education materials (PEMs) in Spanish.MethodsPEMs (n=78) from the American Academy of Orthopaedic Surgery were translated from English to Spanish, using 2 LLMs (GPT-4 and Google Translate). The BLEU methodology was applied to compare AI translations with professionally human-translated PEMs. The Friedman test and Dunn multiple comparisons test were used to statistically quantify differences in translation quality. A readability analysis and feature analysis were subsequently performed to evaluate text simplification success and the impact of English text features on BLEU scores. The capability of an LLM to simplify medical language written in Spanish was also assessed.ResultsAs measured by BLEU scores, GPT-4 showed moderate success in translating PEMs into Spanish but was less successful than Google Translate. Simplified PEMs demonstrated improved readability when compared to original versions (P<.001) but were unable to reach the targeted grade level for simplification. The feature analysis revealed that the total number of syllables and average number of syllables per sentence had the highest impact on BLEU scores. GPT-4 was able to significantly reduce the complexity of medical text written in Spanish (P<.001).ConclusionsAlthough Google Translate outperformed GPT-4 in translation accuracy, LLMs, such as GPT-4, may provide significant utility in translating medical texts into Spanish and simplifying such texts. We recommend considering a dual approach—using Google Translate for translation and GPT-4 for simplification—to improve medical information accessibility and orthopedic surgery education among Spanish-speaking patients.

Machine translation is a computer-based translation process that receives a set of words from a particular humanreadable language as input and outputs a second set of words in the intended human-readable language. Machine translation has assisted linguists and sociologists all across the world. A machine translation model may be statistical, rule-based, or based on neural networks. The Neural Machine Translation (NMT) model was introduced in response to the numerous drawbacks of rule- based and statistical-based machine translation models. This study adds a parallel corpus of Assamese and English and develops the NMT system, a transformer model with a self- attention mechanism. For translation from English to Assamese, the system received a higher bilingual evaluation understudy (BLEU) score of 7.6, and for translation from Assamese to English,a BLEU score of 23.2.

Over the past ten years, neural machine translation (NMT) has seen tremendous growth and is now entering a phase of maturity. Despite being the most popular solution for machine translation (MT), it performs sub-optimally on under-resourced language pairs due to lack of parallel corpora as compared to high-resourced language pairs. The implementation of NMT techniques for under-resourced language pairs is receiving the attention of researchers and has resulted in a significant amount of research for many under-resourced language pairs. In view of the growth of MT, this paper describes a set of practical approaches for investigating MT between Kannada and Tulu. These two languages belong to the family of Dravidian languages and are under-resourced due to lack of tools and resources particularly the parallel corpus for MT. Since there are no parallel corpora for the Kannada-Tulu language pair for MT, this work aims to construct a parallel corpus for this language pair. As manual construction of parallel corpus is laborious, data augmentation is introduced to enhance the size of the parallel corpus along with suitable preprocessing techniques. Different NMT schemes such as recurrent neural network (RNN) baseline, bidirectional recurrent neural network (BiRNN), transformer-based NMT with and without subword tokenization, and statistical machine translation (SMT) models are implemented for MT of Kannada-Tulu and Tulu-Kannada language pairs. Empirical results reveal that the impact of data augmentation increases the bilingual evaluation understudy (BLEU) score of the proposed models. Transformer-based models with subword tokenization outperformed the other models with BLEU scores 41.82 and 40.91 for Kannada-Tulu and Tulu-Kannada MT, respectively.

Recently, semantic communication (SC) has been regarded as one of the most potential paradigms of 6G. Current SC frameworks require the physical layer channel state information (CSI) in order to handle the severe signal distortion induced by channel fading. Since practical CSI cannot be obtained accurately and the overhead of channel estimation cannot be neglected, we therefore propose a generative adversarial network (GAN) based SC framework (Ti-GSC) that doesn’t require CSI. In Ti-GSC, there are two main modules, i.e., an autoencoder-based encoder-decoder module (AEDM) and a GAN-based non-CSI signal distortion suppression (SDS) module (GSDSM), where SDS only relies on learning the syntactic distribution and the semantics of the transmitted data, so no prior information such as CSI is needed by GSDSM. In order to measure signal distortion, a novel loss function is proposed where two terms, i.e., a syntactic distortion loss term and a semantic distortion loss term, are newly added, and a differentiable semantic measurement method is designed based on the intermediate layers of the AEDM decoder. To achieve better training results of Ti-GSC, two training schemes, i.e., the joint optimization based training (JOT) and the alternating optimization based training (AOT) are designed for the proposed Ti-GSC. Experimental results show that JOT is more efficient for Ti-GSC, and Ti-GSC outperforms conventional communication frameworks in terms of bilingual evaluation understudy (BLEU) score in both Rician and Rayleigh fading channels. Moreover, without CSI, the BLEU score achieved by Ti-GSC is about 40% and 62% higher than that achieved by existing SC frameworks in Rician and Rayleigh fading, respectively. Besides, each term of the presented loss function has a great impact on the BLEU performance of Ti-GSC, where in Rician fading syntactic learning has the greatest impact, and in Rayleigh fading, the adversarial learning becomes important.

Applying convolutional attention mechanisms and Human Memory Search for effective English-Urdu translation

BackgroundLarge language models (LLMs) have shown promising abilities in text-based clinical tasks but they do not inherently interpret medical images such as knee radiographs.PurposeTo develop a human-artificial intelligence interactive diagnostic approach, named radiology generative pretrained transformer (RadGPT), aimed at assisting and synergizing with human users for the interpretation of knee radiological images.Materials and MethodsA total of 22 512 knee roentgen ray images and reports were retrieved from Massachusetts General Hospital; 80% of these were used for model training and 10% were used for model testing and validation, respectively. Fifteen diagnostic imaging features (eg, osteoarthritis, effusion, joint space narrowing, osteophyte) were selected to label images based on their high frequency and clinical relevance in the retrieved official reports. Area under the curve scores were calculated for each feature to assess the diagnostic performance. To evaluate the quality of the generated medical text, historical clinical reports were used as the reference text. Several metrics for text generation tasks are applied, including BiLingual Evaluation Understudy, Recall-Oriented Understudy for Gisting Evaluation, Metric for Evaluation of Translation with Explicit Ordering, and Semantic Propositional Image Caption Evaluation.ResultsRadGPT, in collaboration with human users, achieved area under the curve scores ranging from 0.76 for osteonecrosis to 0.91 for arthroplasty across 15 diagnostic categories for knee conditions. Compared with the baseline LLM method, RadGPT achieved higher scores, specifically 0.18 in BiLingual Evaluation Understudy score, 0.30 in Recall-Oriented Understudy for Gisting Evaluation-L, 0.10 in Metric for Evaluation of Translation with Explicit Ordering, and 0.15 in Semantic Propositional Image Caption Evaluation, which is significantly higher than the baseline LLM method, demonstrating good linguistic overlap and clinical consistency with the reference reports.ConclusionRadGPT has achieved advanced results in knee roentgen ray image feature recognition, illustrating the potential of LLMs in medical image interpretation. The study establishes a training protocol for developing artificial intelligence-assisted tools specifically focusing on the diagnosis and interpretation of knee radiological images.

Recent major developments in semantic communication systems stem from integration of deep learning (DL) techniques. Following the discovery of capacity achieving codes, the primary motivation for adopting the semantic approach, which retrieves meaning without requiring an exact reconstruction, is its potential to further conserve resources such as bandwidth and power. In this paper, we propose a novel semantic communication framework for textual data over additive white Gaussian noise (AWGN) channels via DL. Our framework leverages the information bottleneck (IB) principle to balance minimizing bit transmission under wireless channel rate constraints with maximizing semantic information retention. Unlike previous works, we integrate the bilingual evaluation understudy (BLEU) sentence similarity score into the training objective to enhance model performance. In particular, inspired by knowledge distillation, we utilize large language models (LLMs) during training to transfer their knowledge of text semantics into our model. Using IB principle, we train a neural semantic encoder at the transmitter and a neural semantic decoder at the receiver that incorporates into its objective function the rate constraint together with the BLEU score and the knowledge encoded in the soft probabilities produced by the LLM. Through extensive experiments, our proposed framework demonstrates a notable improvement of up to 45% in text semantic similarity compared to state-of-the-art benchmarks operating at the same channel capacity, significantly outperforming traditional communication systems. Moreover, it exhibits robustness to variations in signal-to-noise ratio (SNR) and achieves significant gains across both low and medium SNR regimes.

Neural machine translation is a machine translation system that uses artificial neural networks to identify nonlinear relationships between bilingual sentence pairs. The language English-Kannada pair has met with less attention in the field of Neural Machine Translation (NMT), mostly due to lack of parallel corpora and the complexity of Kannada's linguistic structure. This research proposes a novel NMT model based on Gated Recurrent Units (GRU) and a Hybrid attention mechanism designed specifically for Kannada's morphological complexity and compared with existing models. By adding language-specific preprocessing techniques and employing data augmentation tactics, our model outperforms previous LSTM, GRU and Bi-LSTM-based algorithms in terms of BLEU and METEOR scores. The proposed model with hybrid attention might be a useful substitute in low-resource settings because transformers and other large models require more resources. The comparative study indicates that the proposed methodology improves the translation by 10% increase in Bilingual Evaluation Understudy (BLEU) and accuracy of 80% and achieved a Rouge Score of 0.9.

The complexity of medical texts in Indonesian creates significant barriers to public health literacy, with existing solutions limited by resource constraints. This paper presents a comprehensive text simplification system using the IndoT5 transformer model. Our contributions include the creation of an expert-validated parallel corpus containing $\mathbf{1, 5 0 0}$ complexsimple sentence pairs focused on 50 prevalent diseases from national health insurance data, subsequently expanded to 5,591 samples through data augmentation. The fine-tuned model demonstrates strong performance with BLEU (Bilingual Evaluation Understudy) score of 42.76, SARI (System against Reference for Simplification) score of 58.62, and ROUGE-L (Recall-Oriented Understudy for Gisting Evaluation) score of 0.676. Furthermore, we have deployed a fully functional web application using Flask framework that provides real-time medical text simplification for healthcare providers and the general public. This work establishes both a benchmark dataset and a practical tool for Indonesian medical natural language processing, effectively bridging the health information gap through accessible technology.

Translating the Holy Quran into different languages is a crucial task. It is a daunting one, given the linguistic complexity and rich cultural context of this holy scripture. The recent advancements in technology, especially in artificial intelligence, brought about a pressing need to explore the potential of machine learning-based tools in easing the translation of religious texts, the Holy Quran in particular. This study explores the potential of adopting ChatGPT, a large language model (LLM) powered by machine learning, in English Quranic translation. The study aims to assess the accuracy, adequacy, and cultural sensitivity of ChatGPT's English Quranic translation and identify potential challenges and limitations of using this tool in this domain. The study employed a comparative analysis approach, evaluating ChatGPT's translation of the El Fatiha Chapter against five human-generated translations using BLEU (Bilingual Evaluation Understudy) and METEOR (Metric for Evaluation of Translation with Explicit ORdering) machine translation metrics. The study also utilized an interpretation of the chapter by El Qurtubi to ensure that ChatGPT has the entire linguistic and cultural context of the chapter. The study's findings indicated that ChatGPT's English translation of El Fatiha achieved a moderate level of accuracy and adequacy with mean BLEU and METEOR scores of 0,48 and 0,78, respectively. However, there were areas where ChatGPT's translation could be improved. The use of ML-based tools such as ChatGPT in English Quranic translation offers promising benefits as it can aid translators in the translation process.

This paper aims to build a customized application for multiple language learning. There are various paid platforms and free platforms are available for language learning, however, the user has not given enough customization. Different application programming interfaces namely, speech-to-text, text translation, and text-to-speech are integratedwith multiple options for language selection. To demonstrate and evaluate this customized application, five languages are chosen to execute whereby the accuracy of the outcome can be scrutinized. The outcome is evaluated with native speakers concerning the exactness between the translated text and the translated audio. BLEU (Bilingual Evaluation Understudy) score is done for accessing the accuracy of translation and the Wilcoxon rank test is performed to find out the observed outcome is meeting the expected outcome. The results show that there is a significant difference between expected and observed outcomes for translation accuracy. The architecture of building a customized is discussed in this paper, which enables the user to build their learning app in python.Many real-time scenarios faced by people where a foreign language is required can be solved by this customized application.

With the rapid development of multimodal learning technologies, this work proposes a Future-Aware Multimodal Consistency Translation (FACT) model. This model incorporates future information guidance and multimodal consistency modeling to improve translation quality and enhance language learning efficiency. The model innovatively integrates target future contextual information with a multimodal consistency loss function, effectively capturing the interaction between text and visual information to optimize translation performance. Experimental results show that, in the English-German translation task, the FACT model outperforms the baseline model in both Bilingual Evaluation Understudy (BLEU) and Meteor scores. The model achieves BLEU scores of 41.3, 32.8, and 29.6, and Meteor scores of 58.1, 52.6, and 49.6 on the Multi30K tset16, tset17, and Microsoft Common Objects in Context datasets, respectively, demonstrating its remarkable performance advantages. Significance analysis also verifies this result. Ablation experiments indicate that the future context information supervision function and multimodal consistency loss function are crucial for the model’s performance. Further language learning experiments show that the FACT model significantly outperforms the Transformer model in multiple metrics, encompassing learning efficiency (83.2 words/hour) and translation quality (82.7 points), illustrating its potential in language learning applications. In short, the FACT model holds high application value in multimodal machine translation and language learning. This work provides new ideas and methods, and advances future multimodal translation technology research and applications.

BackgroundIn biomedical text mining, named entity recognition (NER) is an important task used to extract information from biomedical articles. Previously proposed methods for NER are dictionary- or rule-based methods and machine learning approaches. However, these traditional approaches are heavily reliant on large-scale dictionaries, target-specific rules, or well-constructed corpora. These methods to NER have been superseded by the deep learning-based approach that is independent of hand-crafted features. However, although such methods of NER employ additional conditional random fields (CRF) to capture important correlations between neighboring labels, they often do not incorporate all the contextual information from text into the deep learning layers.ResultsWe propose herein an NER system for biomedical entities by incorporating n-grams with bi-directional long short-term memory (BiLSTM) and CRF; this system is referred to as a contextual long short-term memory networks with CRF (CLSTM). We assess the CLSTM model on three corpora: the disease corpus of the National Center for Biotechnology Information (NCBI), the BioCreative II Gene Mention corpus (GM), and the BioCreative V Chemical Disease Relation corpus (CDR). Our framework was compared with several deep learning approaches, such as BiLSTM, BiLSTM with CRF, GRAM-CNN, and BERT. On the NCBI corpus, our model recorded an F-score of 85.68% for the NER of diseases, showing an improvement of 1.50% over previous methods. Moreover, although BERT used transfer learning by incorporating more than 2.5 billion words, our system showed similar performance with BERT with an F-scores of 81.44% for gene NER on the GM corpus and a outperformed F-score of 86.44% for the NER of chemicals and diseases on the CDR corpus. We conclude that our method significantly improves performance on biomedical NER tasks.ConclusionThe proposed approach is robust in recognizing biological entities in text.

The Cambridge University Engineering Department phrase-based statistical machine translation system follows a generative model of translation and is implemented by the composition of component models of translation and movement realised as Weighted Finite State Transducers. Our flexible architecture requires no special purpose decoder and readily handles the large-scale natural language processing demands of state-of-the-art machine translation systems. In this paper we describe the CUED system's participation in the NIST 2008 Arabic-English machine translation evaluation task.