Average Speaker Research Articles

Text-independent voiceprint recognition via compact embedding of dilated deep convolutional neural networks

In order to process speech, most state-of-the-art experimental methods employ convolutional neural networks (CNNs), which operate on a continuous, 1-dimensional (1-D) time stream. In an audio signal, the mel-spectrogram facilitates the representation of attributes of the utterances' in the frequency domain (which corresponds to the speech spectrum). Moreover, for a time-series speaker signal, CNNs are superior to machine or transfer learning models in capturing characteristics from long-form talks. This paper introduces a jump-connected 1-D CNN that employs a combined loss function for speaker recognition. The suggested model uses a 1-D convolutional layer combined with jump connections to extract speaker-specific characteristics; this reduces time-based and frequency-based variability for faster computing. A combined softmax loss, stable L2-norm, and smooth L1-norm loss function guide the proposed compact convolutional neural networks (CCNN) to identify the correct spokesman with improved efficacy. We evaluated the proposed framework using various standard and real-time audio datasets. The experimental findings demonstrate that the proposed CCNN outperforms existing approaches by reducing the equal error rate by 9.02 %. Also, our recommended voiceprint identification model achieves an impressive average speaker recognition rate of 98.76 %. Simultaneously, the reliability of the 1-D CCNN is tested under various conditions. Other fields of study, like language modelling, could employ this approach after some fine-tuning.Relevance of the work: Speaker recognition is an area of interest in which machine learning (ML) and deep learning (DL) schemes, when combined, have the potential to make history in the areas of forensic sciences, automation, and authentication. Using a modest CNN can enhance the identification and verification process by ignoring many issues such as false positives, background noise, and so on. Expanding this process would facilitate raga identification and disease treatment therapies.

Are We Talking about the Same Thing? Modeling Semantic Similarity between Common and Specialized Lexica in WordNet

Specialized languages can activate different sets of semantic features when compared to general language or express concepts through different words according to the domain. The specialized lexicon, i.e., lexical units that denote more specific concepts and knowledge emerging from specific domains, however, co-exists with the common lexicon, i.e., the set of lexical units that denote concepts and knowledge shared by the average speakers, regardless of their specific training or expertise. Communication between specialists and non-specialists can show a big gap between language(s), and therefore lexical units, used by the two groups. However, quite often, semantic and conceptual overlapping between specialized and common lexical units occurs and, in many cases, the specialized and common units refer to close concepts or even point to the same reality. Considering the modeling of meaning in functional lexical resources, this paper puts forth a solution that links common and specialized lexica within the WordNet model framework. We propose a new relation expressing semantic proximity between common and specialized units and define the conditions for its establishment. Besides contributing to the observation and understanding of the process of knowledge specialization and its reflex on the lexicon, the proposed relation allows for the integration of specialized and non-specialized lexicons into a single database, contributing directly to improving communication in specialist/non-specialist contexts, such as teaching–learning situations or health professional-patient interactions, among many others, where code-switching is frequent and necessary.

On commitment to untruthful implicatures

AbstractIn the current debate on the lying-misleading distinction, many theorists distinguish between lying as insincere assertion and misleading through conveying an untruthful implicature. There is growing empirical evidence that average speakers count untruthful implicatures as cases of lying. What matters for them is the (degree) of commitment to an untruthful implicature. Since untruthful conversational implicatures may arise with non-assertions, and untruthful presuppositions are also judged as lying, a realistic conception of lying should aim at a definition of lying that it is able to cover these possibilities. Such a conception, which supports traditional assumptions about the semantics-pragmatics distinction, leads to a commitment-based definition of lying, as recently proposed by a number of authors.

Emotional Speaker Verification Using Novel Modified Capsule Neural Network

Capsule Neural Network (CapsNet) models are regarded as efficient substitutes for convolutional neural networks (CNN) due to their powerful hierarchical representation capability. Nevertheless, CNN endure their inability of recording spatial information in spectrograms. The main constraint of CapsNet is related to the compression method which can be implemented in CNN models but cannot be directly employed in CapsNet. As a result, we propose a novel architecture based on dual-channel long short-term memory compressed CapsNet (DC-LSTM–COMP CapsNet) for speaker verification in emotional as well as stressful talking environments. The proposed approach is perceived as a modified Capsule network that attempts to overcome the limitations that exist within the original CapsNet, as well as in CNN while enhancing the verification performance. The proposed architecture is assessed on four distinct databases. The experimental analysis reveals that the average speaker verification performance is improved in comparison with CNN, the original CapsNet, as well as the conventional classifiers. The proposed algorithm notably achieves the best verification accuracy across the four speech databases. For example, using the Emirati dataset, the average percentage equal error rates (EERs) obtained is 10.50%, based on the proposed architecture which outperforms other deep and classical models.

Verbalization of the Complex Concept Word/Language/Speech from Dual Linguistics Perspective

The verbalization of the concept WORD / LANGUAGE / SPEECH is a complex phenomenon is viewed from the point of view of a new humanitarian branch – dual linguistics, a combination of scientific and naive interpretation of language in the contemporary English lexicography. Language means verbalizing and explaining such interrelated notions as language, word, and speech, constitutes a mosaic-like unity of nominative units fixed in current lexicographic resources. It is proved that the interpretation of this particular phenomenon in the conceptual and linguistic pictures of the world of an average speaker and professional linguist is radically different. The dual linguistics perspective makes it possible to distinguish the two nominative fields of the studied concept: one for the naive and the other for the scientific pictures of the world. Factual data of the study that has been analyzed were drawn, respectively, on the one hand, from general (explanatory) and, on the other hand, special (linguistic) dictionaries of the English language. The primary task of this research is to describe the nuclear zone of the nominative field of the concept WORD/ LANGUAGE / SPEECH, which presupposes the naïve and scientific understanding of language through the prism of English lexicography.

Zamboanga Chavacano: A Survey on Written Practices, Reactions and Impact of its (Standardised) Orthography

Zamboanga Chavacano is the most vibrant Philippine Creole Spanish variety. Since the implementation of the Mother Tongue-Based Multilingual Education (MTB-MLE) program in 2012, it has been taught as a subject and used as a medium of instruction in the public schools of Zamboanga City. Despite the approval of an orthography in 2014, it continues to be written using ad hoc spelling systems influenced by Filipino, English and Spanish to varying degrees. By means of a self-administered questionnaire distributed in social media, this study aims to investigate the level of awareness of the speakers to the orthography, their reactions to it, and show that it may be fairly complex for the average speaker. The results of this survey allowed us to preliminarily identify the contexts in which the language is usually written and devise a suitable corpus of written Chavacano for the study of the spelling errors made by the speakers.

A Robust Approach for Speaker Identification Using Dialect Information

The present research is an effort to enhance the performance of voice processing systems, in our case the speaker identification system (SIS) by addressing the variability caused by the dialectical variations of a language. We present an effective solution to reduce dialect-related variability from voice processing systems. The proposed method minimizes the system’s complexity by reducing search space during the testing process of speaker identification. The speaker is searched from the set of speakers of the identified dialect instead of all the speakers present in system training. The study is conducted on the Pashto language, and the voice data samples are collected from native Pashto speakers of specific regions of Pakistan and Afghanistan where Pashto is spoken with different dialectal variations. The task of speaker identification is achieved with the help of a novel hierarchical framework that works in two steps. In the first step, the speaker’s dialect is identified. For automated dialect identification, the spectral and prosodic features have been used in conjunction with Gaussian mixture model (GMM). In the second step, the speaker is identified using a multilayer perceptron (MLP)-based speaker identification system, which gets aggregated input from the first step, i.e., dialect identification along with prosodic and spectral features. The robustness of the proposed SIS is compared with traditional state-of-the-art methods in the literature. The results show that the proposed framework is better in terms of average speaker recognition accuracy (84.5% identification accuracy) and consumes 39% less time for the identification of speaker.

Conceptual Metaphor Theory analysis of anishi ‘eyes’-based metaphors in Gonja

This article provides a Conceptual Metaphor Theory (CMT) analysis of the metaphorical use of the body part anishi ‘eyes’ in Gonja, a North Guang Niger-Congo language spoken by some 310,000 people predominantly in the Northern and Savannah regions of Ghana. The use of body-part metaphors is an integral part of the daily repertoire of the average Gonja speaker. This article discusses how the anishi ‘eyes’-based expressions are grounded by the biological, psychological and socio-cultural functions of the eye. Prominent anishi ‘eyes’ mappings identified in the language include EMOTIONS/ATTITUDES ARE OBJECTS, ATTITUDES ARE POSSESSIONS, EYE STANDS FOR THE VIEW OR SIGHT, IDEAS AS OBJECTS, IDEAS ARE POSSESSIONS, KNOWLEDGE IS VISION, MANNER FOR ATTITUDE, PAYING ATTENTION IS SEEING, RED EYES ARE SERIOUS, UNAVAILABLE TO KNOWLEDGE IS INVISIBLE OR COVERED, and WET EYES ARE CAREFUL. Regarding the universalist and variationist dichotomy on metaphors, this article argues that, while there is evidence to support the universality of the conceptual metaphors, the importance of cross-cultural distinctiveness cannot be discounted. The data for this work is the product of elicitations from native speaker Gonja students at the University of Education, Winneba, Ghana and complemented with data based on my intuitions as a Safaliba-Gonja bilingual first language speaker. Secondary sources of data include Afari-Twako (2017) and Agyekum (2018).

The Use of ‘woman/female’ and ‘man/male’ as a specific gender markers in Albanian language in light of society changes

Gender issues in Languages are very sensitive and interesting, especially in the societies making effort to empower the presence and the role of women in respective society. Kosovar society is one of the examples of such efforts and debates initiated by women's rights groups and organizations. The situation has changed rapidly in some senses of the use of different linguistics markers regarding the use of respective terms that are considered more appropriate or politically correct than the traditional ones. For example, the terms 'gender' and 'sex' are alternating each-other, not only to denote biological distinctions but also cultural ones. The most rapid changes have been noticed in the use of the term 'woman' instead of the traditional term 'female', which in the perception of the average speakers was not an offensive term for women. But, there is still in use the term 'female' to denote the person of certain gender or sex, especially in informal and social use and in instant messaging. One of the issues identified during the monitoring of media for this paper, was the use of masculine plural to cover also the feminine members of certain groups, especially while naming occupations and this situation is unavoidable because the gender in Albanian has clear grammatical features in the plural as well. To complete this research, I monitored media in different formats (newspapers, online media, social media, radio, and television, in Albanian, Serbian and Macedonian), and also I did a focus group with youngsters on instant messaging platforms.

«Азбучні» полеміки в Україні впродовж ХІХ – початку ХХІ ст.

After numerous discussions, the 2019 «Ukrainian Orthographic Norms» took effect. The innovations proposed in it necessitated turning the pages of the history of Ukrainian graphics and spelling for understanding why «alphabetic wars» in Ukraine continue today and the consequences of such scholarly battles. The purpose of the paper is to give a brief overview of the «spelling wars» in Ukraine in the 19th ‒ early 21st centuries, to expose the causes and consequences of the lingering struggle for the affirmation of specific language norms. In the first half of the 19th century the struggle between the phonetic and etymological traditions of writing was evidenced by O. Pavlovsky’s orthography of the Ukrainian language grammar (1818), the works of M. Maksymovych (1827), and the almanac Rusalka Dnistrova (The Mermaid of the Dniester) in 1837. In the second half of the 19th century new orthographic norms appeared: «kulishivka», «drahomanivka», «zhelekhivka», etc. The early 20th century was crowned by the appearance of «grinchenkivka», the academic publication «The principal rules of the Ukrainian spelling» in 1920, and the «skrypnykivka» spelling in 1928. The reasons of the tragic fate of «skrypnykivka» and the appearance of the Ukrainian spelling systems of the Soviet period in Ukraine in 1933, 1946, and 1960 were found. The attempts to restore the lost national features of the Ukrainian graphics and spelling after the proclamation of Ukraine as an independent state (Orthographic norms of 1989, the drafts of its new editions of 1999 and 2003) proved to be unsuccessful. In the conditions of constant disputes and confrontations the current orthographic code was approved only with some surface changes. There is not a single country in the world with a spelling system the «fate» of which would literally be shot, whose development would be blocked, which would experience political intrusions, and suffer the extermination of national language features. The author of the paper emphasizes that for a long historical period these «alphabetic wars» involved Ukrainians in a negative program of the perception and understanding of spelling rules: one aggression gave rise to a number of others and the internal revolt (primarily – with the introduction of certain features of «skrypnykivka»). According to the author of the article, the norms in the «Ukrainian Orthographic Norms» of 2019 are far from the real life, some formulations are too complicated for an average speaker. 1. We would recommend reducing the number of capitalization rules by listing the concepts that should be written in capital letters. We propose to set some capitalization rules (see §§ 45–48) along with the material on the punctuation of sentences with quotations, remarks, appeals, rubrics, etc. 2. In the media we would like to hear more information about the evolution of language norms and the history of their appearance. Only the wide popularization of the history of spelling and the peaceful solution of spelling problems will help to return to the Ukrainians the distinctive features of orthography. 3. In the search for new grammatical forms, it is important not to push the content into the background. To prevent this from happening, the Ukrainians must learn to think differently, make decisions without fuss, avoid any manifestations of aggression and negative involvement in problems, and seek compromises. In the following scientific investigations, we plan to analyze the punctuation rules of the mother tongue, presented in the current orthographic norms of 2019.

Speaker clustering quality estimation with logistic regression

This paper focuses on estimating the quality of a clustering process. The task is to cluster short speech segments that belong to different speakers. A variety of statistical parameters are estimated from the output of the clustering process. These parameters are used to train a logistic regression to serve as a clustering quality estimation system. In this paper, mean-shift clustering with either a cosine distance or probabilistic linear discriminant analysis (PLDA) score as the similarity measure, as well as stochastic vector quantization (VQ) with cosine distance, are applied in order to cluster the short speaker segments, which are represented by i-vectors. The quality of the clustering is measured using the average cluster purity (ACP), average speaker purity (ASP) and K, which is the geometric mean of ASP and ACP. We show that these measures can be estimated fairly well by applying logistic regression. Moreover, clustering quality may be well estimated even if the logistic regression was trained using parameters derived from a different clustering algorithm. This is very important, as it allows the use of a single quality estimation system, without the need for retraining when the clustering method is changed.Additionally, we showed how the clustering quality estimator could be served as an estimator of the number of clusters. For VQ-based clustering the number of clusters has to be predefined. We perform the clustering with different number of clusters. The best number of clusters is estimated as the clustering that achieved the higher estimation of the K value. We will show that this approach estimate the best number of clusters accurately.

Denominations of Herbs in Těšín Silesia motivated by their therapeutic effects

This study focuses on the names of medicinal plants which are used in the linguistic area of Těšín Silesia and which have been derived from their eff ects known in folk human and veterinary medicine. The etymol-ogy of the names of these plants was transparent to the average speaker of the language and acted as a clear guide on how to use them. It is therefore a suitable source of ethnolinguistic knowledge. We identify four typical types of motivation: the organ or part of body for which the plant is used; the disease against which the plant is used; the eff ect; the person or animal for whose treatment the plant is used. Each type is documented using specifi c examples and aided by analyses of the origin of the name of the plant and its therapeutic use as it is known in Těšín Silesia. In interesting cases, comparison with contemporary Czech and other Slavic languages is also carried out.

Three-stage speaker verification architecture in emotional talking environments

Speaker verification performance in neutral talking environment is usually high, while it is sharply decreased in emotional talking environments. This performance degradation in emotional environments is due to the problem of mismatch between training in neutral environment while testing in emotional environments. In this work, a three-stage speaker verification architecture has been proposed to enhance speaker verification performance in emotional environments. This architecture is comprised of three cascaded stages: gender identification stage followed by an emotion identification stage followed by a speaker verification stage. The proposed framework has been evaluated on two distinct and independent emotional speech datasets: in-house dataset and Emotional Prosody Speech and Transcripts dataset. Our results show that speaker verification based on both gender information and emotion information is superior to each of speaker verification based on gender information only, emotion information only, and neither gender information nor emotion information. The attained average speaker verification performance based on the proposed framework is very alike to that attained in subjective assessment by human listeners.

Emirati-accented speaker identification in each of neutral and shouted talking environments

This work is devoted to capturing Emirati-accented speech database (Arabic United Arab Emirates database) in each of neutral and shouted talking environments in order to study and enhance text-independent Emirati-accented speaker identification performance in shouted environment based on each of First-Order Circular Suprasegmental Hidden Markov Models (CSPHMM1s), Second-Order Circular Suprasegmental Hidden Markov Models (CSPHMM2s), and Third-Order Circular Suprasegmental Hidden Markov Models (CSPHMM3s) as classifiers. In this research, our database was collected from fifty Emirati native speakers (twenty five per gender) uttering eight common Emirati sentences in each of neutral and shouted talking environments. The extracted features of our collected database are called Mel-Frequency Cepstral Coefficients (MFCCs). Our results show that average Emirati-accented speaker identification performance in neutral environment is 94.0%, 95.2%, and 95.9% based on CSPHMM1s, CSPHMM2s, and CSPHMM3s, respectively. On the other hand, the average performance in shouted environment is 51.3%, 55.5%, and 59.3% based, respectively, on CSPHMM1s, CSPHMM2s, and CSPHMM3s. The achieved average speaker identification performance in shouted environment based on CSPHMM3s is very similar to that obtained in subjective assessment by human listeners.

Jak rozwalić system – populistyczny dyskurs polityczny na przykładzie wypowiedzi Pawła Kukiza

<p>Analiza wypowiedzi Pawła Kukiza pozwala stwierdzić, że przypisywany politykowi populizm widoczny jest zarówno w treściowej warstwie wypowiedzi, jak i jej językowym kształcie. Wystąpienia sejmowe i posty zamieszczane na portalu społecznościowym Facebook zawierają wszystkie cechy konstytutywne dla populizmu, czyli centralnie usytuowany lud w kreowanej przez polityka wizji świata, który przedstawiany jest zawsze w opozycji do elit lub obcych, a także symplifikację językowego obrazu świata (w tym wartości), ponadto obecność lidera występującego z woli ludu. Językowy kształt wypowiedzi polityka odwołuje się do języka potocznego, bliskiego przeciętnemu użytkownikowi języka, dzięki czemu wypowiedzi lidera Kukiz’15 są emocjonalne, wyraziście wartościują opisywaną rzeczywistość, także przez stereotypy i metafory. Tak ukształtowany sposób komunikowania się z odbiorcami tworzy spójną wizję z przyjmowanymi przez polityka rolami, w które wchodzi, przywołując archetypy wojownika, chłopaka z sąsiedztwa, nie wyzbywając się nigdy pozycji dominującej.</p>

Orthophonic errors/clippings in speeches delivered by prominent Polish politicians (2009-2016)

This paper tackles the issue of incorrect abbreviating original words in the Polish language used by the most prominent Polish political figures of the recent years. Orthophony as such is extremely frequently violated in everyday language. The language of Polish broadcast media, contrary to what it was like a few decades ago, seems to follow the trends observed in the speech of ordinary citizens of the country. Moreover, official speeches delivered by major politicians appear to reflect the nonchalance of average speakers. In this article a preliminary analysis of the issue of language decadence at the heights of power is proposed.

Assisting non-expert speakers of under-resourced languages in assigning stems and inflectional paradigms to new word entries of morphological dictionaries

This paper presents a new method with which to assist individuals with no background in linguistics to create monolingual dictionaries such as those used by the morphological analysers of many natural language processing applications. The involvement of non-expert users is especially critical for under-resourced languages which either lack or cannot afford the recruitment of a skilled workforce. Adding a word to a morphological dictionary usually requires identifying its stem along with the inflection paradigm that can be used in order to generate all the word forms of the new entry. Our method works under the assumption that the average speakers of a language can successfully answer the polar question “is x a valid form of the word w to be inserted?”, where x represents tentative alternative (inflected) forms of the new word w. The experiments show that with a small number of polar questions the correct stem and paradigm can be obtained from non-experts with high success rates. We study the impact of different heuristic and probabilistic approaches on the actual number of questions.

The process of constructing ontological meaning based on criminal law verbs

This study intends to account for the process involved in the construction of the conceptual meaning of verbs (#EVENTS) directly related to legal aspects of terrorism and organized crime based on the evidence provided by the Globalcrimeterm Corpus and the consistent application of specific criteria for term extraction. The selected 49 concepts have eventually been integrated in the Core Ontology of FunGramKB (Functional Grammar Knowledge Base), a knowledge base which is founded on the principles of deep semantics and is also aimed at the computational development of the Lexical Constructional Model (www.fungramkb.com). To achieve this purpose, key phases of the COHERENT methodology (Periñán Pascual & Mairal Usón 2011) are followed, particularly those which involve the modelling, subsumption and hierarchisation of the aforementioned verbal concepts. The final outcome of this research shows that most of the apparently specialised conceptual units should eventually be included in the Core Ontology instead of the specific Globalcrimeterm Subontology, due to the fact that the semantic content of their corresponding lexical units can be found in widely used learner`s dictionaries and, consequently, this conceptual information is not only shared by the experts in the field but also by the layperson and the average speaker of the language.

Speaker Identification in a Shouted Talking Environment Based on Novel Third-Order Circular Suprasegmental Hidden Markov Models

It is well known that speaker identification yields very high performance in a neutral talking environment, on the other hand, the performance has been sharply declined in a shouted talking environment. This work aims at proposing, implementing, and evaluating novel Third-Order Circular Suprasegmental Hidden Markov Models (CSPHMM3s) to improve the low performance of text-independent speaker identification in a shouted talking environment. CSPHMM3s possess combined characteristics of: Circular Hidden Markov Models (CHMMs), Third-Order Hidden Markov Models (HMM3s), and Suprasegmental Hidden Markov Models (SPHMMs). Our results show that CSPHMM3s are superior to each of: First-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM1s), Second-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM2s), Third-Order Left-to-Right Suprasegmental Hidden Markov Models (LTRSPHMM3s), First-Order Circular Suprasegmental Hidden Markov Models (CSPHMM1s), and Second-Order Circular Suprasegmental Hidden Markov Models (CSPHMM2s) in a shouted talking environment. Using our collected speech database, average speaker identification performance in a shouted talking environment based on LTRSPHMM1s, LTRSPHMM2s, LTRSPHMM3s, CSPHMM1s, CSPHMM2s, and CSPHMM3s is 74.6%, 78.4%, 81.7%, 78.7%, 83.4%, and 85.8%, respectively. Speaker identification performance that has been achieved based on CSPHMM3s is close to that attained based on subjective assessment by human listeners.

An analytical model to evaluate the cocktail party effect in restaurant dining rooms: A case study

This paper proposes the use of a simplified analytical model to evaluate acoustic conditions in restaurant dining rooms required for ensuring the intelligibility of conversations. The model is useful for design applications and is suitable for evaluating the maximum number of speakers present in a restaurant room in order to ensure intelligibility of conversations taking place at each table in the presence of background noise caused by conversations at other tables. The maximum number of speakers is studied in relation to the sound level difference between useful and disturbing sound, sound absorption of the room, and the average speaker–listener distance. The model is applied to the case of a dining hall in a multipurpose centre, which is currently in the planning stage.