One of the main tasks in the field of natural language processing (NLP) is the analysis of affective states (sentiment and emotional) based on written text, and attempts have improved dramatically in ...recent years. However, in studies on the Arabic language, machine learning or deep learning algorithms were utilised to analyse sentiment and emotion more often than current pre-trained language models. Additionally, further pre-training the language model on specific tasks (i.e., within-task and cross-task adaptation) has not yet been investigated for Arabic in general, and for the sentiment and emotion task in particular. In this paper, we adapt a BERT-based Arabic pretrained language model for the sentiment and emotion tasks by further pre-training it on a sentiment and emotion corpus. Hence, we developed five new Arabic models: QST, QSR, QSRT, QE3, and QE6. Five sentiment and two emotion datasets spanning both small- and large-resource settings were used to evaluate the developed models. The adaptation approaches significantly enhanced the performance of seven Arabic sentiment and emotion datasets. The developed models showed excellent improvements over the sentiment and emotion datasets, which ranged from 0.15–4.71%.
Understanding the public’s diverse linguistic expressions about rainfall and flood provides a basis for flood disaster studies and enhances linguistic and cultural awareness. However, existing ...research tends to overlook linguistic complexity, potentially leading to bias. In this study, we introduce a novel algorithm capturing rainfall and flood-related expressions, considering the relationship between precipitation observations and linguistics expressions. Analyzing 210 million social media microblogs from 2017, we identified 594 keywords, 20 times more than usual manually created bag-of-words. Utilizing Large Language Model, we categorized these keywords into rainfall, flood, and other related terms. Semantic features of these keywords were analyzed from the viewpoint of popularity, credibility, time delay, and part-of-speech, finding rainfall-related terms most common-used, flood-related keywords often more time delayed than precipitation, and notable differences in part-of-speech across categories. We also assessed spatial characteristics from keyword and city-centric perspectives, revealing that 49.5% of the keywords have significant spatial correlation with differing median centers, reflecting regional variations. Large and disaster-impacted cities show the richest expression diversity for rainfall and flood-related terms.
This paper reports on activites at LIMSI over the last few years directed at the transcription of broadcast news data. We describe our development work in moving from laboratory read speech data to ...real-world or `found' speech data in preparation for the DARPA evaluations on this task from 1996 to 1999. Two main problems needed to be addressed to deal with the continuous flow of inhomogenous data. These concern the varied acoustic nature of the signal (signal quality, environmental and transmission noise, music) and different linguistic styles (prepared and spontaneous speech on a wide range of topics, spoken by a large variety of speakers).
The problem of partitioning the continuous stream of data is addressed using an iterative segmentation and clustering algorithm with Gaussian mixtures. The speech recognizer makes use of continuous density HMMs with Gaussian mixture for acoustic modeling and 4-gram statistics estimated on large text corpora. Word recognition is performed in multiple passes, where current hypotheses are used for cluster-based acoustic model adaptation prior to the next decoding pass. The overall word transcription error of the LIMSI evaluation systems were 27.1% (Nov96, partitioned test data), 18.3% (Nov97, unpartitioned data), 13.6% (Nov98, unpartitioned data) and 17.1% (Fall99, unpartitioned data with computation time under 10× real-time).
Dieser Artikel berichtet über die Tätigkeiten am LIMSI während der letzten Jahren mit dem Ziel der Spracherkennung von Nachrichtensendungen. Wir beschreiben unsere Forschungsarbeiten der Portierung von unter Laborbedingungen gelesener Sprache zu natürlicher freier Sprache während der Vorbereitung der ARPA Nov96, Nov97, Nov98 und Nov99 Evaluierungen. Zur Bearbeitung des kontinuierlichen Stroms von inhomogenen Daten sind zwei grundlegende Probleme zu lösen.
Diese betreffen einerseits die unregelmässige akustische Natur des Signals (Signalqualität, Hintergrund- und Übertragungsrauschen, Musik,…) und andererseits die unterschiedlichen linguistischen Stile (vorbereitete oder spontane Sprache, eine große Themenvielfalt und viele unterschiedliche Sprecher).
Der kontinuierliche Audiostrom wird mit Hilfe eines iterativen Segmentierungs- und Klusterungsalgorithmus auf der Basis von Gauss Mischverteilungen partioniert. Das Spracherkennungssystem verwendet HMMs mit kontinuierlichen Gauss Mischverteilungen zur akustischen Modellierung und 4-gram Statistiken, welche mit Hilfe großer Textsammlungen geschätzt wurden. Die Worterkennung erfolgt in mehreren Phasen, wobei die Wortgraphen nach und nach mit Hilfe akustischer Modellanpassung verbessert werden. Die Wordfehlerrate von LIMSI's Spracherkennungssytemen beträgt 27,1% (Nov96, segmentierte Testdaten), 18,3% (Nov98, unsegmentierte Testdaten) und 17,1% (Herbst 99, unsegmentierte Testdaten und Ausführungszeiten von weniger als 10-facher Echtzeit).
Cet article présente les travaux effectués au LIMSI pour le développement d'un système de traitement automatique d'informations radio et télédiffusées. Partant d'un système de transcription de textes lus, nous décrivons les adaptations qui ont été nécessaires pour le traitement d'un flux audio continu et de données dites “trouvées”. Ces développements ont été validés dans le cadre des évaluations ARPA BN (Nov96, Nov97, Nov98 et Dec99). Les principales difficultés posées par ce type de données sont liées à leur nature hétérogène, qu'il s'agisse de changements de nature acoustique (environnement, communication, musique) ou de nature linguistique (styles d'élocution, diversités des sujets et des locuteurs).
La partition du flux continu est effectuée de manière itérative, par un algorithme de segmentation–agglomération reposant sur des mélanges de Gaussiennes. Le système de reconnaissance utilise des modèles de Markov cachés à densités continues pour la modélisation acoustique, et des statistiques 4-grammes de mots estimées sur un grand corpus de textes et de parole transcrite pour modèle de langage. La transcription en mots est obtenue en plusieurs passes de décodage, où les hypothèses intermédiaires sont utilisées pour adapter les modèles acoustiques. Les taux d'erreur obtenues avec différentes versions de ce système lors des évaluations ARPA sont 27,1% (Nov96 avec partition manuelle), 18,3% (Nov97), 13,6% (Nov98) et 17,1% (Dec99, moins de 10 fois le temps réel).
Research in corpus-driven Automatic Speech Recognition (ASR) is advancing rapidly towards building a robust Large Vocabulary Continuous Speech Recognition (LVCSR) system. Under-resourced languages ...like Bangla require benchmarking large corpora for more research on LVCSR to tackle their limitations and avoid the biased results. In this paper, a publicly published large-scale Bangladeshi Bangla speech corpus is used to implement deep Convolutional Neural Network (CNN) based model and Recurrent Neural Network (RNN) based model with Connectionist Temporal Classification (CTC) loss function for Bangla LVCSR. In experimental evaluations, we find that CNN-based architecture yields superior results over the RNN-based approach. This study also emphasizes assessing the quality of an open-source large-scale Bangladeshi Bangla speech corpus and investigating the effect of the various high-order N-gram Language Models (LM) on a morphologically rich language Bangla. We achieve 36.12% word error rate (WER) using CNN-based acoustic model and 13.93% WER using beam search decoding with 5-gram LM. The findings demonstrate by far the state-of-the-art performance of any Bangla LVCSR system on a specific benchmarked large corpus.
Dysarthric speech has several pathological characteristics, such as discontinuous pronunciation, uncontrolled volume, slow speech, explosive pronunciation, improper pauses, excessive nasal sounds, ...and air-flow noise during pronunciation, which differ from healthy speech. Automatic speech recognition (ASR) can be very helpful for speakers with dysarthria. Our research aims to provide a scoping review of ASR for dysarthric speech, covering papers in this field from 1990 to 2022. Our survey found that the development of research studies about the acoustic features and acoustic models of dysarthric speech is nearly synchronous. During the 2010s, deep learning technologies were widely applied to improve the performance of ASR systems. In the era of deep learning, many advanced methods (such as convolutional neural networks, deep neural networks, and recurrent neural networks) are being applied to design acoustic models and lexical and language models for dysarthric-speech-recognition tasks. Deep learning methods are also used to extract acoustic features from dysarthric speech. Additionally, this scoping review found that speaker-dependent problems seriously limit the generalization applicability of the acoustic model. The scarce available speech data cannot satisfy the amount required to train models using big data.
Connectionist Temporal Classification (CTC) loss has become widely used in sequence modeling tasks such as Automatic Speech Recognition (ASR) and Handwritten Text Recognition (HTR) due to its ease of ...use. Recent sequence models that incorporate CTC loss have been focusing on speed by removing recurrent structures, hence losing important context information. This paper presents extensive studies of Contextualized Connectionist Temporal Classification (CCTC) framework, which induces prediction dependencies in non-recurrent and non-autoregressive neural networks for sequence modeling. CCTC allows the model to implicitly learn the language model by predicting neighboring labels via multi-task learning. Experiments on ASR and HTR tasks in two different languages show that CCTC models offer improvements over CTC models by 2.2-8.4% relative without incurring extra inference costs. We have also found that higher order of context information can potentially help the model produce better predictions.
In this paper we address the automatic generation of literary phrases in Spanish. We propose a model for text generation based on statistical algorithms and Shallow Parsing. We present preliminary ...results that are quite encouraging.
Despite the freedom of information and the development of various open data repositories, the access to legal information to general audience remains hindered due to the difficulty of understanding ...and interpreting it. In this paper we aim at employing modern language models to extract the most important information from legal documents and structure this information in a knowledge graph. This knowledge graph can later be used to retrieve information and answer legal question. To evaluate the performance of different models we formalize the task as event extraction and manually annotate 133 instances. We evaluate two models: GRIT and Text2Event. The latter model achieves a better score of ~ 0.8 F1 score for identifying legal classes and 0.5 F1 score for identifying roles in legal relations. We demonstrate how the produced legal knowledge graph could be exploited with 2 example use cases. Finally, we annotate the whole Workers’ Statute using the fine-tuned Text2Event model and publish the results in an open repository.
•Features embedded the information of word history could improve modeling accuracy.•Word-clustering with decision trees could increase modeling accuracy and handle OOVs.•Subspace Gaussian mixture ...models would alleviate the sparseness of language modeling.•Models of a family of Gaussian distribution have been already developed deeply.
This paper focuses on adaptable continuous space language modeling approach of combining longer context information of recurrent neural network (RNN) with adaptation ability of subspace Gaussian mixture model (SGMM) which has been widely used in acoustic modeling for automatic speech recognition (ASR).
In large vocabulary continuous speech recognition (LVCSR) it is a challenging problem to construct language models that can capture the longer context information of words and ensure generalization and adaptation ability. Recently, language modeling based on RNN and its variants have been broadly studied in this field.
The goal of our approach is to obtain the history feature vectors of a word with longer context information and model every word by subspace Gaussian mixture model such as Tandem system used in acoustic modeling for ASR. Also, it is to apply fMLLR adaptation method, which is widely used in SGMM based acoustic modeling, for adaptation of subspace Gaussian mixture based language model (SGMLM).
After fMLLR adaptation, SGMLMs based on Top-Down and Bottom-Up obtain WERs of 5.70 % and 6.01%, which are better than 4.15% and 4.61% of that without adaptation, respectively. Also, with fMLLR adaptation, Top-Down and Bottom-Up based SGMLMs yield absolute word error rate reduction of 1.48%, 1.02% and a relative perplexity reduction of 10.02%, 6.46% compared to RNNLM without adaptation, respectively.
Most characters in the Chinese and Japanese languages are ideographic compound characters composedof subcharacter elements arranged in a planar manner. Token-based and image-based subcharacter-level ...modelshave been proposed to leverage the subcharacter elements. However, on the one hand, the conventional tokenbasedsubcharacter-level models are blind to the planar structural information; on the other hand, the image-basedmodels are weak with respect to ideographic characters that share similar shapes but have different meanings.These characteristics motivate us to explore non-image-based methods to encode the planar structural informationof characters. In this paper, we propose and discuss a method to encode the planar structural information by learningembeddings of the categories of the structure types. Our proposed model adds the structure embeddings to theconventional subcharacter embeddings and position embeddings before they are input into the encoder. In this way,the model learns the planar structural and positional information and retains the uniqueness of each character. Weevaluated the method in a text classification task. In the experiment, the embeddings were encoded by a CNN encoder,and then the encoded vectors were input into an LSTM classifier to classify product reviews as positive or negative.We compared the proposed model with models that use only the subcharacter embeddings, the structure embeddingor the position embeddings as well as with the conventional models in previous works. The results show that addingboth structure embeddings and position embeddings leads to more rich and representative features and better fittingon the dataset. The proposed method results in at best 1:8% better recall, 0:63% better F-score, and 0:55% better accuracy on the testing datasets compared to previous methods.
PDF
