Search results for: collecting speech emotion corpus

Authors: A. Shoiynbek, K. Kozhakhmet, P. Menezes, D. Kuanyshbay, D. Bayazitov

Abstract:

Speech emotion recognition has received increasing research interest all through current years. There was used emotional speech that was collected under controlled conditions in most research work. Actors imitating and artificially producing emotions in front of a microphone noted those records. There are four issues related to that approach, namely, (1) emotions are not natural, and it means that machines are learning to recognize fake emotions. (2) Emotions are very limited by quantity and poor in their variety of speaking. (3) There is language dependency on SER. (4) Consequently, each time when researchers want to start work with SER, they need to find a good emotional database on their language. In this paper, we propose the approach to create an automatic tool for speech emotion extraction based on facial emotion recognition and describe the sequence of actions of the proposed approach. One of the first objectives of the sequence of actions is a speech detection issue. The paper gives a detailed description of the speech detection model based on a fully connected deep neural network for Kazakh and Russian languages. Despite the high results in speech detection for Kazakh and Russian, the described process is suitable for any language. To illustrate the working capacity of the developed model, we have performed an analysis of speech detection and extraction from real tasks.

Keywords: deep neural networks, speech detection, speech emotion recognition, Mel-frequency cepstrum coefficients, collecting speech emotion corpus, collecting speech emotion dataset, Kazakh speech dataset

Procedia PDF Downloads 66

Authors: L. Kamandulytė-Merfeldienė

Abstract:

The paper deals with the main issues of methodology of the Corpus of Spoken Lithuanian which was started to be developed in 2006. At present, the corpus consists of 300,000 grammatically annotated word forms. The creation of the corpus consists of three main stages: collecting the data, the transcription of the recorded data, and the grammatical annotation. Collecting the data was based on the principles of balance and naturality. The recorded speech was transcribed according to the CHAT requirements of CHILDES. The transcripts were double-checked and annotated grammatically using CHILDES. The development of the Corpus of Spoken Lithuanian has led to the constant increase in studies on spontaneous communication, and various papers have dealt with a distribution of parts of speech, use of different grammatical forms, variation of inflectional paradigms, distribution of fillers, syntactic functions of adjectives, the mean length of utterances.

Keywords: CHILDES, corpus of spoken Lithuanian, grammatical annotation, grammatical disambiguation, lexicon, Lithuanian

Procedia PDF Downloads 204

Authors: Tomasz Sapiński, Dorota Kamińska, Adam Pelikant, Egils Avots, Cagri Ozcinar, Gholamreza Anbarjafari

Abstract:

People express emotions through different modalities. Integration of verbal and non-verbal communication channels creates a system in which the message is easier to understand. Expanding the focus to several expression forms can facilitate research on emotion recognition as well as human-machine interaction. In this article, the authors present a Polish emotional database composed of three modalities: facial expressions, body movement and gestures, and speech. The corpora contains recordings registered in studio conditions, acted out by 16 professional actors (8 male and 8 female). The data is labeled with six basic emotions categories, according to Ekman’s emotion categories. To check the quality of performance, all recordings are evaluated by experts and volunteers. The database is available to academic community and might be useful in the study on audio-visual emotion recognition.

Keywords: body movement, emotion recognition, emotional corpus, facial expressions, gestures, multimodal database, speech

Procedia PDF Downloads 316

Authors: Esha Banerjee, Atul Kumar Ojha, Girish Nath Jha

Abstract:

This study aims to develop an intonation labeled database for Hindi. Although no single standard for prosody labeling exists in Hindi, researchers in the past have employed perceptual and statistical methods in literature to draw inferences about the behavior of prosody patterns in Hindi. Based on such existing research and largely agreed upon intonational theories in Hindi, this study attempts to develop a manually annotated prosodic corpus of Hindi speech data, which can be used for training speech models for natural-sounding speech in the future. 100 sentences ( 500 words) each for declarative and interrogative types have been labeled using Praat.

Keywords: speech dataset, Hindi, intonation, labeled corpus

Procedia PDF Downloads 156

Authors: Stefanie Rukavina, Sascha Gruss, Steffen Walter, Holger Hoffmann, Harald C. Traue

Abstract:

OPEN-EmoRecII is an open multimodal corpus with experimentally induced emotions. In the first half of the experiment, emotions were induced with standardized picture material and in the second half during a human-computer interaction (HCI), realized with a wizard-of-oz design. The induced emotions are based on the dimensional theory of emotions (valence, arousal and dominance). These emotional sequences - recorded with multimodal data (mimic reactions, speech, audio and physiological reactions) during a naturalistic-like HCI-environment one can improve classification methods on a multimodal level. This database is the result of an HCI-experiment, for which 30 subjects in total agreed to a publication of their data including the video material for research purposes. The now available open corpus contains sensory signal of: video, audio, physiology (SCL, respiration, BVP, EMG Corrugator supercilii, EMG Zygomaticus Major) and mimic annotations.

Keywords: open multimodal emotion corpus, annotated labels, intelligent interaction

Procedia PDF Downloads 376

Authors: Bubai Maji, Monorama Swain

Abstract:

Speech is considered an essential and most natural medium for the interaction between machines and humans. However, extracting effective features for speech emotion recognition (SER) is remains challenging. The present studies show that the temporal information captured but high-level temporal-feature learning is yet to be investigated. In this paper, we present an efficient novel method using the Self-attention (SA) mechanism in a combination of Convolutional Neural Network (CNN) and Bi-directional Gated Recurrent Unit (Bi-GRU) network to learn high-level temporal-feature. In order to further enhance the representation of the high-level temporal-feature, we integrate a Bi-GRU output with learnable weights features by SA, and improve the performance. We evaluate our proposed method on our created SITB-OSED and IEMOCAP databases. We report that the experimental results of our proposed method achieve state-of-the-art performance on both databases.

Keywords: Bi-GRU, 1D-CNNs, self-attention, speech emotion recognition

Procedia PDF Downloads 84

Authors: John Lorenzo Bautista, Yoon-Joong Kim

Abstract:

This paper presents the development of a Filipino speech recognition tool using the HTK System. The system was trained from a subset of the Filipino Speech Corpus developed by the DSP Laboratory of the University of the Philippines-Diliman. The speech corpus was both used in training and testing the system by estimating the parameters for phonetic HMM-based (Hidden-Markov Model) acoustic models. Experiments on different mixture-weights were incorporated in the study. The phoneme-level word-based recognition of a 5-state HMM resulted in an average accuracy rate of 80.13 for a single-Gaussian mixture model, 81.13 after implementing a phoneme-alignment, and 87.19 for the increased Gaussian-mixture weight model. The highest accuracy rate of 88.70% was obtained from a 5-state model with 6 Gaussian mixtures.

Keywords: Filipino language, Hidden Markov Model, HTK system, speech recognition

Procedia PDF Downloads 435

Authors: Leah Lee

Abstract:

This paper examines the correlation between deep learning models and emotions with noises to see whether or not noises mask emotions. The deep learning models used are plain convolutional neural networks (CNN), auto-encoder, long short-term memory (LSTM), and Visual Geometry Group-16 (VGG-16). Emotion datasets used are Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), Crowd-sourced Emotional Multimodal Actors Dataset (CREMA-D), Toronto Emotional Speech Set (TESS), and Surrey Audio-Visual Expressed Emotion (SAVEE). To make it four times bigger, audio set files, stretch, and pitch augmentations are utilized. From the augmented datasets, five different features are extracted for inputs of the models. There are eight different emotions to be classified. Noise variations are white noise, dog barking, and cough sounds. The variation in the signal-to-noise ratio (SNR) is 0, 20, and 40. In summation, per a deep learning model, nine different sets with noise and SNR variations and just augmented audio files without any noises will be used in the experiment. To compare the results of the deep learning models, the accuracy and receiver operating characteristic (ROC) are checked.

Keywords: auto-encoder, convolutional neural networks, long short-term memory, speech emotion recognition, visual geometry group-16

Procedia PDF Downloads 35

Authors: Takayuki Konishi, Kakeru Yazawa, Mariko Kondo

Abstract:

This study examines developmental change in the production of epenthetic vowels by Japanese learners of English in relation to acquisition of L2 English speech rhythm. Seventy-two Japanese learners of English in the J-AESOP corpus were divided into lower- and higher-level learners according to their proficiency score and the frequency of vowel epenthesis. Three learners were excluded because no vowel epenthesis was observed in their utterances. The analysis of their read English speech data showed no statistical difference between lower- and higher-level learners, implying the absence of any developmental change in durations of epenthetic vowels. This result, together with the findings of previous studies, will be discussed in relation to the transfer of L1 phonology and manifestation of L2 English rhythm.

Keywords: vowel epenthesis, Japanese learners of English, L2 speech corpus, speech rhythm

Procedia PDF Downloads 236

Authors: Wujian Han

Abstract:

In the process of human interaction and communication, speech act verbs are considered to be the most active component and the main means for information transmission, and are also taken as an indication of the structure of linguistic behavior. The theoretical value and practical significance of such everyday built-in metalanguage have long been recognized. This paper, which is part of a bigger study, is aimed to provide useful insights for a more precise and systematic application to speech act verbs translation between English and Chinese, especially with regard to the degree to which generic integrity is maintained in the practice of translation of legal documents. In this study, the corpus, i.e. Chinese legal texts and their English translations, English legal texts, ordinary Chinese texts, and ordinary English texts, serve as a testing ground for examining contrastively the usage of English and Chinese directive speech act verbs in legal genre. The scope of this paper is relatively wide and essentially covers all directive speech act verbs which are used in ordinary English and Chinese, such as order, command, request, prohibit, threat, advice, warn and permit. The researcher, by combining the corpus methodology with a contrastive perspective, explored a range of characteristics of English and Chinese directive speech act verbs including their semantic, syntactic and pragmatic features, and then contrasted them in a structured way. It has been found that there are similarities between English and Chinese directive speech act verbs in legal genre, such as similar semantic components between English speech act verbs and their translation equivalents in Chinese, formal and accurate usage of English and Chinese directive speech act verbs in legal contexts. But notable differences have been identified in areas of difference between their usage in the original Chinese and English legal texts such as valency patterns and frequency of occurrences. For example, the subjects of some directive speech act verbs are very frequently omitted in Chinese legal texts, but this is not the case in English legal texts. One of the practicable methods to achieve adequacy and conciseness in speech act verb translation from Chinese into English in legal genre is to repeat the subjects or the message with discrepancy, and vice versa. In addition, translation effects such as overuse and underuse of certain directive speech act verbs are also found in the translated English texts compared to the original English texts. Legal texts constitute a particularly valuable material for speech act verb study. Building up such a contrastive picture of the Chinese and English speech act verbs in legal language would yield results of value and interest to legal translators and students of language for legal purposes and have practical application to legal translation between English and Chinese.

Keywords: contrastive analysis, corpus-based, directive speech act verbs, legal texts, translation between English and Chinese

Procedia PDF Downloads 441

Authors: R. Ajgou, S. Sbaa, S. Ghendir, A. Chemsa, A. Taleb-Ahmed

Abstract:

The speech enhancement algorithm is to improve speech quality. In this paper, we review some speech enhancement methods and we evaluated their performance based on Perceptual Evaluation of Speech Quality scores (PESQ, ITU-T P.862). All method was evaluated in presence of different kind of noise using TIMIT database and NOIZEUS noisy speech corpus.. The noise was taken from the AURORA database and includes suburban train noise, babble, car, exhibition hall, restaurant, street, airport and train station noise. Simulation results showed improved performance of speech enhancement for Tracking of non-stationary noise approach in comparison with various methods in terms of PESQ measure. Moreover, we have evaluated the effects of the speech enhancement technique on Speaker Identification system based on autoregressive (AR) model and Mel-frequency Cepstral coefficients (MFCC).

Keywords: speech enhancement, pesq, speaker recognition, MFCC

Procedia PDF Downloads 383

Authors: Rahul Mohite, Lahcen Ouarbya

Abstract:

This paper focuses on the interplay between humans and computer systems and the ability of these systems to understand and respond to human emotions, including non-verbal communication. Current emotion recognition systems are based solely on either facial or verbal expressions. The limitation of these systems is that it requires large training data sets. The paper proposes a system for recognizing human emotions that combines both speech and emotion recognition. The system utilizes advanced techniques such as deep learning and image recognition to identify facial expressions and comprehend emotions. The results show that the proposed system, based on the combination of facial expression and speech, outperforms existing ones, which are based solely either on facial or verbal expressions. The proposed system detects human emotion with an accuracy of 86%, whereas the existing systems have an accuracy of 70% using verbal expression only and 76% using facial expression only. In this paper, the increasing significance and demand for facial recognition technology in emotion recognition are also discussed.

Keywords: facial reputation, expression reputation, deep gaining knowledge of, photo reputation, facial technology, sign processing, photo type

Procedia PDF Downloads 72

Authors: Yishu Gong, Liangliang Yang, Jianyu Zhang, Zhengyu Chen, Sihong He, Xusheng Zhang, Wei Zhang

Abstract:

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that affects millions of people worldwide and is characterized by cognitive decline and behavioral changes. People living with Alzheimer’s disease often find it hard to complete routine tasks. However, there are limited objective assessments that aim to quantify the difficulty of certain tasks for AD patients compared to non-AD people. In this study, we propose to use speech emotion recognition (SER), especially the frustration level, as a potential biomarker for quantifying the difficulty patients experience when describing a picture. We build an SER model using data from the IEMOCAP dataset and apply the model to the DementiaBank data to detect the AD/non-AD group difference and perform longitudinal analysis to track the AD disease progression. Our results show that the frustration level detected from the SER model can possibly be used as a cost-effective tool for objective tracking of AD progression in addition to the Mini-Mental State Examination (MMSE) score.

Keywords: Alzheimer’s disease, speech emotion recognition, longitudinal biomarker, machine learning

Procedia PDF Downloads 76

Authors: P. S. Jagadeesh Kumar, Yang Yung, Wenli Hu

Abstract:

Speech emotion classification is a dominant research field in finding a sturdy and profligate classifier appropriate for different real-life applications. This effort accentuates on classifying different emotions from speech signal quarried from the features related to pitch, formants, energy contours, jitter, shimmer, spectral, perceptual and temporal features. Tensor deep stacking neural networks were supported to examine the factors that influence the classification success rate. Facial electromyography signals were composed of several forms of focuses in a controlled atmosphere by means of audio-visual stimuli. Proficient facial electromyography signals were pre-processed using moving average filter, and a set of arithmetical features were excavated. Extracted features were mapped into consistent emotions using bilinear mapping. With facial electromyography signals, a database comprising diverse emotions will be exposed with a suitable fine-tuning of features and training data. A success rate of 92% can be attained deprived of increasing the system connivance and the computation time for sorting diverse emotional states.

Keywords: speech emotion classification, tensor deep stacking neural networks, facial electromyography, bilinear mapping, audio-visual stimuli

Procedia PDF Downloads 211

Authors: Shaode Yu, Jiajian Meng, Bing Zhu, Hang Yu, Qiurui Sun

Abstract:

Speech emotion recognition (SER) is for recognizing human subjective emotions through audio data in-depth analysis. From speech audios, how to comprehensively extract emotional information and how to effectively fuse extracted features remain challenging. This paper presents a dual-stream SER framework that embraces both full training and transfer learning of different networks for thorough feature encoding. Besides, a plug-and-play cross-attention fusion (CAF) module is implemented for the valid integration of the dual-stream encoder output. The effectiveness of the proposed CAF module is compared to the other three fusion modules (feature summation, feature concatenation, and feature-wise linear modulation) on two databases (RAVDESS and IEMO-CAP) using different dual-stream encoders (full training network, DPCNN or TextRCNN; transfer learning network, HuBERT or Wav2Vec2). Experimental results suggest that the CAF module can effectively reconcile conflicts between features from different encoders and outperform the other three feature fusion modules on the SER task. In the future, the plug-and-play CAF module can be extended for multi-branch feature fusion, and the dual-stream SER framework can be widened for multi-stream data representation to improve the recognition performance and generalization capacity.

Keywords: speech emotion recognition, cross-attention fusion, dual-stream, pre-trained

Procedia PDF Downloads 35

Authors: M. C. Padma, R. J. Prathibha

Abstract:

Parts of speech tagging is the process of assigning appropriate parts of speech tags to the words in a given text. The critical or crucial information needed for tagging a word come from its internal structure rather from its neighboring words. The internal structure of a word comprises of its morphological features and grammatical information. This paper presents a morpheme based parts of speech tagger for Kannada language. This proposed work uses hierarchical tag set for assigning tags. The system is tested on some Kannada words taken from EMILLE corpus. Experimental result shows that the performance of the proposed system is above 90%.

Keywords: hierarchical tag set, morphological analyzer, natural language processing, paradigms, parts of speech

Procedia PDF Downloads 258

Authors: Seham El Kareh, Miramar Etman

Abstract:

Recently, detecting liars and extracting features which distinguish them from truth-tellers have been the focus of a wide range of disciplines. To the author’s best knowledge, most of the work has been done on facial expressions and body gestures but only few works have been done on the language used by both liars and truth-tellers. This paper sheds light on four axes. The first axis copes with building an audio corpus for deceptive and truthful speech for Egyptian Arabic speakers. The second axis focuses on examining the human perception of lies and proving our need for computational linguistic-based methods to extract features which characterize truthful and deceptive speech. The third axis is concerned with building a linguistic analysis program that could extract from the corpus the inter- and intra-linguistic cues for deceptive and truthful speech. The program built here is based on selected categories from the Linguistic Inquiry and Word Count program. Our results demonstrated that Egyptian Arabic speakers on one hand preferred to use first-person pronouns and present tense compared to the past tense when lying and their lies lacked of second-person pronouns, and on the other hand, when telling the truth, they preferred to use the verbs related to motion and the nouns related to time. The results also showed that there is a need for bigger data to prove the significance of words related to emotions and numbers.

Keywords: Egyptian Arabic corpus, computational analysis, deceptive features, forensic linguistics, human perception, truthful features

Procedia PDF Downloads 176

Authors: Robert Long, Hiroaki Watanabe

Abstract:

This longitudinal study examines the grammatical errors of Japanese university students’ dialogues with a native speaker over an academic year. The L2 interactions of 15 Japanese speakers were taken from the JUSFC2018 corpus (April/May 2018) and the JUSFC2019 corpus (January/February). The corpora were based on a self-introduction monologue and a three-question dialogue; however, this study examines the grammatical accuracy found in the dialogues. Research questions focused on a possible significant difference in grammatical accuracy from the first interview session in 2018 and the second one the following year, specifically regarding errors in clauses per 100 words, global errors and local errors, and with specific errors related to parts of speech. The investigation also focused on which forms showed the least improvement or had worsened? Descriptive statistics showed that error-free clauses/errors per 100 words decreased slightly while clauses with errors/100 words increased by one clause. Global errors showed a significant decline, while local errors increased from 97 to 158 errors. For errors related to parts of speech, a t-test confirmed there was a significant difference between the two speech corpora with more error frequency occurring in the 2019 corpus. This data highlights the difficulty in having students self-edit themselves.

Keywords: clause analysis, global vs. local errors, grammatical accuracy, L2 output, longitudinal study

Procedia PDF Downloads 89

Authors: Aishwarya Ravindra Fursule, Shruti Kshirsagar

Abstract:

In the intersection of artificial intelligence and human-centered computing, this paper delves into speech emotion recognition (SER). It presents a comparative analysis of machine learning models such as K-Nearest Neighbors (KNN),logistic regression, support vector machines (SVM), decision trees, ensemble classifiers, and random forests, applied to SER. The research employs four datasets: Crema D, SAVEE, TESS, and RAVDESS. It focuses on extracting salient audio signal features like Zero Crossing Rate (ZCR), Chroma_stft, Mel Frequency Cepstral Coefficients (MFCC), root mean square (RMS) value, and MelSpectogram. These features are used to train and evaluate the models’ ability to recognize eight types of emotions from speech: happy, sad, neutral, angry, calm, disgust, fear, and surprise. Among the models, the Random Forest algorithm demonstrated superior performance, achieving approximately 79% accuracy. This suggests its suitability for SER within the parameters of this study. The research contributes to SER by showcasing the effectiveness of various machine learning algorithms and feature extraction techniques. The findings hold promise for the development of more precise emotion recognition systems in the future. This abstract provides a succinct overview of the paper’s content, methods, and results.

Keywords: comparison, ML classifiers, KNN, decision tree, SVM, random forest, logistic regression, ensemble classifiers

Procedia PDF Downloads 3

Authors: Mirela-Gabriela Bratu

Abstract:

The main objective of this study is to highlight the way in which emotions are constructed discursively in the French written press, more particularly in the titles of informative articles. To achieve this objective, we will begin the study with the theoretical part, which aims to capture the characteristics of journalistic discourse, to which we will add clues of emotions that we will identify in the titles of the articles. The approach is based on the empirical results from the analysis of the articles published on the earthquake that took place on August 24, 2016, in Italy, as described by two French national daily newspapers: Le Monde and Le Point. The corpus submitted to the analysis contains thirty-seven titles, published between August 24, 2016, and August 24, 2017. If the textual content of the speech offers information respecting the grammatical standards and following the presentation conventions, the choice of words can touch the reader, so the journalist must add other means than mastering of the language to create emotion. This study aims to highlight the strategies, such as rhetorical figures, the tenses, or factual data, used by journalists to create emotions for the readers. We also try, thanks to the study of the articles which were published for several days relating to the same event, to emphasize whether we can speak or not of the dissipation of emotion and the catastrophic side as the event fades away in time. The theoretical framework is offered by works on rhetorical strategies (Perelman, 1992; Amossi, 2000; Charaudeau, 2000) and on the study of emotions (Plantin, 1997, 1998, 2004; Tetu, 2004).

Keywords: disaster, earthquake, emotion, feeling

Procedia PDF Downloads 104

Authors: Mahboobeh Sadeghi, Zahra Izadi Khah, Mansour Hakim Javadi, Masoud Gholamali Lavasani

Abstract:

Background: Since there is a relation between psychological and physiological factors, the aim of this study was to examine the effect of Emotion Regulation training on cognitive emotion regulation problem in patients with Multiple Sclerosis(MS) Method: In a randomized clinical trial thirty patients diagnosed with Multiple Sclerosis referred to state welfare organization were selected. The sample group was randomized into either an experimental group or a nonintervention control group. The subjects participated in 75-minute treatment sessions held three times a week for 4weeks (12 sessions). All 30 individuals were administered with Cognitive Emotion Regulation questionnaire (CERQ). Participants completed the questionnaire in pretest and post-test. Data obtained from the questionnaire was analyzed using Mancova. Results: Emotion Regulation significantly decreased the Cognitive Emotion Regulation problems patients with Multiple sclerosis (p < 0.001). Conclusions: Emotion Regulation can be used for the treatment of cognitive-emotion regulation problem in Multiple sclerosis.

Keywords: Multiple Sclerosis, cognitive-emotion regulation, emotion regulation, MS

Procedia PDF Downloads 419

Authors: Ericson O. Alieto, Aprillete C. Devanadera

Abstract:

The main objective of the study is to determine the dominant speech acts in five commemorative speeches of President Duterte. This study employed Speech Act Theory and Discourse analysis to determine how the speech acts features connote the pragmatic meaning of Duterte’s speeches. Identifying the speech acts is significant in elucidating the underlying message or the pragmatic meaning of the speeches. From the 713 sentences or utterances from the speeches, assertive with 208 occurrences from the corpus or 29% is the dominant speech acts. It was followed by expressive with 177 or 25% occurrences, directive accounts for 152 or 15% occurrences. While commisive accounts for 104 or 15% occurrences and declarative got the lowest percentage of occurrences with 72 or 10% only. These sentences when uttered by Duterte carry a certain power of language to move or influence people. Thus, the present study shows the fundamental message perceived by the listeners. Moreover, the frequent use of assertive and expressive not only explains the pragmatic message of the speeches but also reflects the personality of President Duterte.

Keywords: commemorative speech, discourse analysis, duterte, pragmatics

Procedia PDF Downloads 248

Authors: Francielle A. Vargas, Thiago A. S. Pardo

Abstract:

According to the standard philosophical definition, lying is saying something that you believe to be false with the intent to deceive. For deception detection, the FBI trains its agents in a technique named statement analysis, which attempts to detect deception based on parts of speech (i.e., linguistics style). This method is employed in interrogations, where the suspects are first asked to make a written statement. In this poster, we model false statements using linguistics style. In order to achieve this, we methodically analyze linguistic features in a corpus of fake news in the Portuguese language. The results show that they present substantial lexical, syntactic and semantic variations, as well as punctuation and emotion distinctions.

Keywords: deception detection, linguistics style, computational linguistics, natural language processing

Procedia PDF Downloads 176

Authors: Shamm Kidane, Ibrahim Abdella, Fitsum Gaim, Simon Mulugeta, Sirak Asmerom, Natnael Ambasager, Yoel Ghebrihiwot

Abstract:

The call for natural language processing (NLP) systems for low-resource languages has become more apparent than ever in the past few years, with the arduous challenges still present in preparing such systems. This paper presents an improved dataset version of the Nagaoka Tigrinya Corpus for Parts-of-Speech (POS) classification system in the Tigrinya language. The size of the initial Nagaoka dataset was incremented, totaling the new tagged corpus to 118K tokens, which comprised the 12 basic POS annotations used previously. The additional content was also annotated manually in a stringent manner, followed similar rules to the former dataset and was formatted in CONLL format. The system made use of the novel approach in NLP tasks and use of the monolingually pre-trained TiELECTRA, TiBERT and TiRoBERTa transformer models. The highest achieved score is an impressive weighted F1-score of 94.2%, which surpassed the previous systems by a significant measure. The system will prove useful in the progress of NLP-related tasks for Tigrinya and similarly related low-resource languages with room for cross-referencing higher-resource languages.

Keywords: Tigrinya POS corpus, TiBERT, TiRoBERTa, conditional random fields

Procedia PDF Downloads 49

Authors: Vadivel Ayyasamy

Abstract:

The present approach deals with the identification of Emotions and classification of Emotional patterns at Phrase-level with respect to Positive and Negative Orientation. The proposed approach considers emotion triggered terms, its co-occurrence terms and also associated sentences for recognizing emotions. The proposed approach uses Part of Speech Tagging and Emotion Actifiers for classification. Here sentence patterns are broken into phrases and Neuro-Fuzzy model is used to classify which results in 16 patterns of emotional phrases. Suitable intensities are assigned for capturing the degree of emotion contents that exist in semantics of patterns. These emotional phrases are assigned weights which supports in deciding the Positive and Negative Orientation of emotions. The approach uses web documents for experimental purpose and the proposed classification approach performs well and achieves good F-Scores.

Keywords: emotions, sentences, phrases, classification, patterns, fuzzy, positive orientation, negative orientation

Procedia PDF Downloads 347

Authors: Chunying Fang, Haifeng Li, Lin Ma, Mancai Zhang

Abstract:

Speech intelligibility assessment is an important measure to evaluate the functional outcomes of surgical and non-surgical treatment, speech therapy and rehabilitation. The assessment of pathological speech plays an important role in assisting the experts. Pathological speech usually is non-stationary and mutational, in this paper, we describe a multi-granularity combined feature schemes, and which is optimized by hierarchical visual method. First of all, the difference granularity level pathological features are extracted which are BAFS (Basic acoustics feature set), local spectral characteristics MSCC (Mel s-transform cepstrum coefficients) and nonlinear dynamic characteristics based on chaotic analysis. Latterly, radar chart and F-score are proposed to optimize the features by the hierarchical visual fusion. The feature set could be optimized from 526 to 96-dimensions.The experimental results denote that new features by support vector machine (SVM) has the best performance, with a recognition rate of 84.4% on NKI-CCRT corpus. The proposed method is thus approved to be effective and reliable for pathological speech intelligibility evaluation.

Keywords: pathological speech, multi-granularity feature, MSCC (Mel s-transform cepstrum coefficients), F-score, radar chart

Procedia PDF Downloads 254

Authors: Anup Anand Deshmukh, Catherine Soladie, Renaud Seguier

Abstract:

Emotion plays a key role in many applications like healthcare, to gather patients’ emotional behavior. Unlike typical ASR (Automated Speech Recognition) problems which focus on 'what was said', it is equally important to understand 'how it was said.' There are certain emotions which are given more importance due to their effectiveness in understanding human feelings. In this paper, we propose an approach that models human stress from audio signals. The research challenge in speech emotion detection is finding the appropriate set of acoustic features corresponding to an emotion. Another difficulty lies in defining the very meaning of emotion and being able to categorize it in a precise manner. Supervised Machine Learning models, including state of the art Deep Learning classification methods, rely on the availability of clean and labelled data. One of the problems in affective computation is the limited amount of annotated data. The existing labelled emotions datasets are highly subjective to the perception of the annotator. We address the first issue of feature selection by exploiting the use of traditional MFCC (Mel-Frequency Cepstral Coefficients) features in Convolutional Neural Network. Our proposed Emo-CNN (Emotion-CNN) architecture treats speech representations in a manner similar to how CNN’s treat images in a vision problem. Our experiments show that Emo-CNN consistently and significantly outperforms the popular existing methods over multiple datasets. It achieves 90.2% categorical accuracy on the Emo-DB dataset. We claim that Emo-CNN is robust to speaker variations and environmental distortions. The proposed approach achieves 85.5% speaker-dependant categorical accuracy for SAVEE (Surrey Audio-Visual Expressed Emotion) dataset, beating the existing CNN based approach by 10.2%. To tackle the second problem of subjectivity in stress labels, we use Lovheim’s cube, which is a 3-dimensional projection of emotions. Monoamine neurotransmitters are a type of chemical messengers in the brain that transmits signals on perceiving emotions. The cube aims at explaining the relationship between these neurotransmitters and the positions of emotions in 3D space. The learnt emotion representations from the Emo-CNN are mapped to the cube using three component PCA (Principal Component Analysis) which is then used to model human stress. This proposed approach not only circumvents the need for labelled stress data but also complies with the psychological theory of emotions given by Lovheim’s cube. We believe that this work is the first step towards creating a connection between Artificial Intelligence and the chemistry of human emotions.

Keywords: deep learning, brain chemistry, emotion perception, Lovheim's cube

Procedia PDF Downloads 120

Authors: Punnoose A. K.

Abstract:

This paper presents an approach for identifying noisy speech recording using a multi-layer perception (MLP) trained to predict phonemes from acoustic features. Characteristics of the MLP posteriors are explored for clean speech and noisy speech at the frame level. Appropriate density functions are used to fit the softmax probability of the clean and noisy speech. A function that takes into account the ratio of the softmax probability density of noisy speech to clean speech is formulated. These phoneme independent scoring is weighted using a phoneme-specific weightage to make the scoring more robust. Simple thresholding is used to identify the noisy speech recording from the clean speech recordings. The approach is benchmarked on standard databases, with a focus on precision.

Keywords: noisy speech identification, speech pre-processing, noise robustness, feature engineering

Procedia PDF Downloads 87

Authors: Mohamed S. Negm, Waleed S. Mandour

Abstract:

The present study aims primarily at investigating words of peace (lexemes of peace) in the formal speeches of the Egyptian president Abdulfattah El-Sisi in a two-year span of time, from 2018 to 2019. This paper attempts to shed light not only on the contextual use of the antonyms, war and peace, but also it underpins quantitative analysis through the current methods of corpus linguistics. As such, the researchers have deployed a corpus-based approach in collecting, encoding, and processing 30 presidential speeches over the stated period (23,411 words and 25,541 tokens in total). Further, semantic fields and collocational networkzs are identified and compared statistically. Results have shown a significant propensity of adopting peace, including its relevant collocation network, textually and therefore, ideationally, at the expense of war concept which in most cases surfaces euphemistically through the noun conflict. The president has not justified the action of war with an honorable cause or a valid reason. Such results, so far, have indicated a positive sociopolitical mindset the Egyptian president possesses and moreover, reveal national and international fair dealing on arising issues.

Keywords: CADS, collocation network, corpus linguistics, critical discourse analysis

Procedia PDF Downloads 111

Authors: Maha Saliba Foster

Abstract:

Speech perception has been recognized as a bi-sensory process involving the auditory and visual channels. Compared to the auditory modality, the contribution of the visual signal to speech perception is not very well understood. Studying how the visual modality affects speech recognition can have pedagogical implications in second language learning, as well as clinical application in speech therapy. The current investigation explores the potential effect of speech visual cues on the perception of Arabic emphatics (AEs). The corpus consists of 36 minimal pairs each containing two contrasting consonants, an AE versus a non-emphatic (NE). Movies of four Lebanese speakers were edited to allow perceivers to have partial view of facial regions: lips only, lips-cheeks, lips-chin, lips-cheeks-chin, lips-cheeks-chin-neck. In the absence of any auditory information and relying solely on visual speech, perceivers were above chance at correctly identifying AEs or NEs across vowel contexts; moreover, the models were able to predict the probability of perceivers’ accuracy in identifying some of the COIs produced by certain speakers; additionally, results showed an overlap between the measurements selected by the computer and those selected by human perceivers. The lack of significant face effect on the perception of AEs seems to point to the lips, present in all of the videos, as the most important and often sufficient facial feature for emphasis recognition. Future investigations will aim at refining the analyses of visual cues used by perceivers by using Principal Component Analysis and including time evolution of facial feature measurements.

Keywords: Arabic emphatics, machine learning, speech perception, visual speech perception

Procedia PDF Downloads 267