Diacritical Marks Research Articles

Morphological, syntactic and diacritics rules for automatic diacritization of Arabic sentences

The diacritical marks of Arabic language are characters other than letters and are in the majority of cases absent from Arab writings. This paper presents a hybrid system for automatic diacritization of Arabic sentences combining linguistic rules and statistical treatments. The used approach is based on four stages. The first phase consists of a morphological analysis using the second version of the morphological analyzer Alkhalil Morpho Sys. Morphosyntactic outputs from this step are used in the second phase to eliminate invalid word transitions according to the syntactic rules. Then, the system used in the third stage is a discrete hidden Markov model and Viterbi algorithm to determine the most probable diacritized sentence. The unseen transitions in the training corpus are processed using smoothing techniques. Finally, the last step deals with words not analyzed by Alkhalil analyzer, for which we use statistical treatments based on the letters. The word error rate of our system is around 2.58% if we ignore the diacritic of the last letter of the word and around 6.28% when this diacritic is taken into account.

Arabic speech synthesis and diacritic recognition

Text-to-speech system (TTS), known also as speech synthesizer, is one of the important technology in the last years due to the expanding field of applications. Several works on speech synthesizer have been made on English and French, whereas many other languages, including Arabic, have been recently taken into consideration. The area of Arabic speech synthesis has not sufficient progress and it is still in its first stage with a low speech quality. In fact, speech synthesis systems face several problems (e.g. speech quality, articulatory effect, etc.). Different methods were proposed to solve these issues, such as the use of large and different unit sizes. This method is mainly implemented with the concatenative approach to improve the speech quality and several works have proved its effectiveness. This paper presents an efficient Arabic TTS system based on statistical parametric approach and non-uniform units speech synthesis. Our system includes a diacritization engine. Modern Arabic text is written without mention the vowels, called also diacritic marks. Unfortunately, these marks are very important to define the right pronunciation of the text which explains the incorporation of the diacritization engine to our system. In this work, we propose a simple approach based on deep neural networks. Deep neural networks are trained to directly predict the diacritic marks and to predict the spectral and prosodic parameters. Furthermore, we propose a new simple stacked neural network approach to improve the accuracy of the acoustic models. Experimental results show that our diacritization system allows the generation of full diacritized text with high precision and our synthesis system produces high-quality speech.

‘Students get bogged down’: how religious Israeli elementary teachers view problems and solutions in Bible teaching

Bible teachers in contemporary society confront serious problems related to the nature of the biblical text and the socio-cultural context of their teaching. This study, based on semi-structured interviews, examines the problems that five expert religious Israeli elementary school teachers encounter in their teaching and the solutions they employ. Our findings show two major domains of pedagogic issues: unfamiliar biblical linguistics and problematic content. Teachers reported student difficulties in understanding biblical Hebrew. Problematic content includes irrelevant topics, emotionally laden material, and age inappropriate issues. Linguistic solutions relied on reading comprehension techniques and use of features specific to Bible reading such as diacritical marks. Regarding content issues, teachers were motivated by faith in the sanctity of the text to find effective solutions. These include selectivity, reinterpretation using homiletic tools, a holistic understanding and contextualising the narrative. Though teachers felt ill-prepared by their pre-service training in dealing with these challenges, they demonstrated resilience in their solution-oriented pedagogy. These findings suggest attention to mentoring and professional development, and to the creation of a community of practice to support teachers’ dealing with the ongoing challenges in their teaching.

Zen Koans by Steven Heine (review)

No! We have Buddha Nature.” In addition, there are cross-references between the historical material of kōan texts and contemporary Western literature and poetry (e.g.: “The significance of the Mu Kōan might also be examined in light of other instances of Western cultural expressions of nothingness or the de-centeredness of the universe in the schools of thought of American Transcendentalism, Dadaism, and Existentialism,” p. 71), and ranging to “Whitehead’s holistic metaphysics” and “Einstein’s quantum physics.” Whereas cross-references to more contemporary discourses might be useful for addressing a Western readership, these references remain too fragmentary and blurry, and do not really add anything substantial to the understanding of the topic discussed. As a consequence, when reading the book, at times it was difficult for me to figure out for what audience the book was written: although providing a wealth of interesting material, the treatment of it is often too unfocused and “jumpy” for a specialized reader. On the other hand, a more general audience might have difficulties in dealing with the overload of methodological reflections and—occasionally—very detailed discussions of kōan features . As such, I regard the publication as a brave attempt of “performative scholarship” by a very accomplished scholar who got a bit too much carried away by the subject he was dealing with (sometimes even adopting the rhetorical features of the kōan texts discussed), and occasionally overloading it with too ambitious methodological endeavors. Based on my subjective reading of the publication, this approach was not quite convincing for me. CHRISTOPH ANDERL Ghent University STEVEN HEINE, Zen Koans. Dimensions of Asian Spirituality. Honolulu: University of Hawai‘i Press, 2014. ix, 195 pp. US $17 (pb). ISBN 978-0-8248-3974-1 Even though Steven Heine’s Zen Koans seems not to be written for specialist readers with a background in Chinese religions—it provides no diacritical marks for Japanese words like kōan (gong’an 公案) and no Sinitic logographs, for example—this short monograph is an important book that ought to be read by anyone with an interest in the literary and philosophical history of Zen Buddhism. But let me make one point very clear: Steven Heine knows Zen. By Zen, I mean the Zen traditions in Japan (Sōtōshū 曹洞宗, Rinzaishū 臨濟宗, and Ō bakushū 黄檗宗), the thriving discourse of Zen studies in Japan (at universities such as Komazawa 駒澤 大學 and Hanazono 花園大學), and Zen literature—especially collections of kōans, or public cases. In Zen Koans, Heine uses more than 26 of the 48 cases of the seminal Gateless Gate (Wumen guan, Mumonkan 無門關) collection from China—in addition to thirty-two references to other normative Zen texts (see pp. 189-192)—to provide the target audience of non-specialist readers with a condensed overview of the philosophical, historical, intellectual, and institutional contexts of a kōan tradition in East Asia (China, Japan, Korea). Heine accomplishes this feat in two chapters and in only 70 pages. The remainder of this short monograph is where Heine seems to distill decades of prolific study of Zen, chiefly focused on either Sōtōshū Zen in Japan or Dōgen 道元 (1200-1253), into another 90 pages and three chapters. Heine can do this with relative ease because he has published 82 BOOK REVIEWS (or coedited, with Dale S. Wright) more than a dozen books on the topic of kōans, and another seven on Dōgen studies. The last chapter of Zen Koans, chapter 6 (“Whither Koans, or Wither? Current Trends and Possible New Directions,” pp. 160-188), is perhaps the most thought provoking of this densely packed monograph , because Heine addresses at least three rousing issues intimately connected to Zen (and kōans) in the 21st century. First, what are we to make of the fact that Japan remains the stimulus for academic research on Zen while South Korean Sŏn and Chinese Chan communities, respectively, grow by leaps and bounds in rapidly changing societies where corruption, both within and beyond the monastic grounds (e.g., gambling in Korea, p. 171), is as much of an issue as practice-based matters related to reading kōans? Second, how are kōans read today by East...

Bridging the digital divide in Sri Lanka: some challenges and opportunities in using Sinhala in ICT

The "digital divide" is the gap in technology usage and access. The digital divide has been investigated by scholars [1] and policy makers [2] mainly as an economy-specific issue that permeates the population across all demographic profiles, such as income, gender, age, education, race, and region, but not specific to the languages of different communities. The lack of native language driven ICT is a major conducive factor in digital divide.Sinhala writing system used in Sri Lanka is a syllabic writing system derived from Brahmi which consist of vowels, consonants, diacritical marks and special symbols constructs. Several of these constructs are combined to form complex ligatures. The total number of different glyphs is almost close to 2300 in Sinhala language. Thus, all computer equipments that support Sinhala language needs to support a greater degree of complexity in both display and printing with near minimal changes to the keyboard or the input systems. In this paper we discuss (1) historical background of the Sinhala writing system, (2) Sinhala scripts’ characteristics and complexities and illustrate (3) how Sinhala computing technology has evolved over the last quarter century. Major steps are marked by the design of character code standards as a corner stone of whole architecture for text processing. A case described in this article of “Digital Inclusion” shows how small communities of non-Roman script users can connect to the Romanized system dominated cyberspace.

THE LINGUISTIC LANDSCAPE OF PRINT ADVERTISEMENTS IN DAUGAVPILS: 1920–1930

<p><em>The theme described in the title is connected with public advertising texts from two informational sources: newspapers printed in Daugavpils („Daugavas Vēstnesis” ‘The Daugava Herald’, „Daugavas Vārds” ‘The Daugava’s Word’, „Latgales Ziņas” ‘Latgalian News’ and „</em><em>Latgales Vēstnesis”</em><em> ‘The Latgale Herald”) and the linguistic landscape of the city, which characterizes the public information space. Commercial discourse is essential to this space, as a large part of public texts have the representation and promotion of establishments, companies and societies as a primary goal, in addition to the exhibition of offered goods and services.</em></p><p><em>The aim of the article is to define and characterize from the perspective of linguistic landscape the tools and techniques used to represent businesses and establishments in Latvian print advertisements in the 1920s and 30s. In fulfillment of this goal, content analysis and the diachronic linguistic landscape approach has been used for data analysis and interpretation (Backhaus 2005, Pavlenko 2010, Pavlenko, Mullen 2015, Pošeiko 2015). For summarization of obtained results, the descriptive method has been used.</em></p><p><em>Latvia is characterized in the interwar period by a unified language policy – including policies with mechanisms for the management of specific languages – highlighting the role of the Latvian language as the state language in the organization of public life and in nationalist ideology, and facilitating its use in all sociolinguistic functions. However, the interwar period in Latvia also marks the beginning of a period of Westernization – especially in the economic and cultural spheres – detectible in cinema, theater and concert posters; print advertisements for shops and consumer services, and business names in the urban environment.</em></p><em>During this period newspapers were printed in Latvian, but some papers, calendars and journals were printed in Latgalian, Russian and Polish. Company names, advertisements and partially-legible posters are visible in period photographs of the linguistic landscape. Advertising information at the beginning of the 1920s is only to be found in Russian, or with bilingual Russian-Latvian texts. Monolingual language signs in Latvian – noticeably missing diacritic marks and appropriate word endings – only begin to be seen from the 1930s.</em>

The terms designating diacritical Marks above leters and punctuation Marks in the Serbian manuscript Miscelany at the Austrian National Library in Viena (cod.Slav.93)

The study presents the previously unknown list that contains terms designating diacritical marks above letters and punctuation marks, written out on fol. 111v of the Liturgical Miscellany (early 18th century, Vienna, Austrian National Library, Cod. slav. 93). The list offers the most comprehensive record of these marks in Serbian manuscript heritage. It may be assumed with great probability that the scribe of the Miscellany adopted these terms either from the Primer by Theophan Prokopovich, which was published on several occasions between 1724 and 1734, or another, earlier primer in the Russo-Slavonic language. This is indicated by the fact that this type of compendium contains almost identical lists of prosodic and punctuation marks.

Brouillons d’un Baiser: Premiers pas vers “Finnegans Wake,” by James Joyce

Reviewed by: Brouillons d’un Baiser: Premiers pas vers “Finnegans Wake,” by James Joyce Terence Killeen (bio) Brouillons d’un Baiser: Premiers pas vers “Finnegans Wake,” by James Joyce, edited by Daniel Ferrer and translated by Marie Darrieussecq. Paris: NRF/Gallimard, 2014. 133pp. €14.90. A kiss is just a kiss? Not always, and certainly not in the case of this book, where the ramifications of a kiss eventually contribute to the genesis of the daunting complexity of Finnegans Wake. At the core of Brouillons d’un baiser (best translated as “sketches of a kiss”) is a set of five brief texts by Joyce, all with links, as argued by editor Daniel Ferrer, to the first kiss of Tristan and Isolde: a kiss that resonates through much western literature and music, especially that of Richard Wagner—as well as throughout Finnegans Wake. Ferrer’s introduction traces Joyce’s interest in the Tristan and Isolde legend: by March 1923, it formed the basis for several fragments that are part of the very earliest writing leading to the Wake. The sketches are given in English, and in French translation by the distinguished French novelist Marie Darrieussecq, who also contributes [End Page 727] a preface to the text. It describes the way she confronted the task of translation and also how a trip to Dublin, where she saw some of the original manuscripts and visited the Joyce Tower at Sandycove, helped with her work. While I am not the ideal person to assess the translations (for obvious reasons), to me, they seem to respond very creatively to the linguistic inventiveness, the fluctuating tonal registers, and especially the humor of these extraordinary pieces. Two and a half of the sketches come from the manuscripts acquired in 2006 by the National Library of Ireland,1 with the remainder from manuscripts held in the British Library2 and known to Joyce scholars for a long time.3 The rest of the work contains an introduction and notes by the French Joycean Ferrer. His text provides an essential background for understanding the place of these pieces in Joyce’s ongoing Wake project; his notes provide details of their provenance, references to various relevant notebook units, and comments on the placing of particular passages. This book is a “reader’s edition”; the sketches are not festooned with diacritical marks to indicate insertions, deletions, alterations, or the like. Rather, they read smoothly, but Ferrer does indicate in his notes many interlinear additions and other such alterations.4 The only other work that has published a “reader’s edition” of sketches by Joyce, in any language, is Danis Rose’s Finn’s Hotel.5 The two books, however, are not directly comparable: Rose prints ten early sketches—about all there are—in the belief that these form a separate work from the Wake itself, originally planned by Joyce and to be called Finn’s Hotel. Ferrer has no such view: “Il est donc tout à fait abusif de présenter ces brouillons comme des œuvres.” (“It is therefore quite improper to treat these sketches as works”—56 n2). Instead, Ferrer is quite clear that the studies are indeed just preliminary to the Wake and have no independent existence of their own. His interest is confined to the Tristan legend and its offshoot in the characters of the Four Old Men and Mamalujo; he does not include the other sketches.6 In that case, is there a justification for publishing them separately from the work itself? The answer to this question is yes; the sketches do have a special quality of their own, distinct in nature from the finished text. For one thing, there is a far greater sense of personality, even of character, about these personae—Issy, Tristan, the Four Old Men—than in the final, published version of the Wake. Second, there is a pronounced difference in the tone of the pieces as compared to the final text. The first sketch is written in a childish voice, enumerating the many virtues of Issy, who is presented as a very young girl, given to good works. She encounters Saint Dympna, herself a victim of incestuous desires, converts a dragon to “the...

“By her unveil’d each horrid crime appears”: Authorship, Text, and Subtext in Phillis Wheatley’s Variants Poems

In 1773, Phillis Wheatley’s Poems on Various Subjects, Religious and Moral appeared print. Ever since the publication of her book of neoclassic verse, the African-born poet has been a controversial figure in American History. At the center of the controversy is the question of whether or not the mother of the African American literary tradition criticized slavery. While some scholars have denounced Wheatley for not addressing the institution; others argue that her work represented a subtle critique. Ironically, missing in this discourse are the poet’s diacritical marks that underscores not only the power of words to mean, but also subversive readings—both of which are the focus of this essay.

Introduction: Art as parodic practice

Difference through repetition marks the domain of our renewed questioning in this special issue. Every utterance – be it diacritical mark, word, paint stroke or musical bar – every iteration of mode, or genre, means dialogue with previous iterations (Bakhtin 1986: 77). This runs the gamut of transformation from sampling, re-mash, pastiche, and travesty, through subversive laughter and extreme semiotic violence, right to revolutionary upheaval – stages through which parody spaces its becoming.

Hindi-Kannada Named Entity Transliteration: Issues and Possible Solutions

Indian languages belong to four language families, namely, the Indo-Aryan, Dravidian, Tibeto-Burman and the Austro- Asiatic. Hindi and Kannada belong to Indo-Aryan and Dravidian family respectively and are evolved from the ancient Brahmi script and have a common phonetic structure. But the Named Entity writing convention is different due to dialectic influence, language specific rules, and other factors. Due to this, the Named Entity Transliteration from Hindi to Kannada and vice versa is not one to one character mapping. This introduces many problems in Machine Translation (MT), Cross Lingual Information Retrieval (CLIR) and Parallel corpus creation between Hindi and Kannada. The paper discusses the Named Entity Transliteration issues encountered between Hindi and Kannada during the parallel corpora creation from Hindi to Kannada for the Indian Language Corpus Initiative (ILCI) project. In this paper, we discuss cases of no exact equivalence character between Hindi and Kannada, multiple mappings, diacritic marks, loan words and language specific transliteration issues in detail and propose the possible solution to resolve the problem. At implementation level, one may make use of either Finite-State Transducers (FST) or Regular Expressions

Design and Implementation of Shahmukhi Spell Checker

Spellchecker is a software tool that identifies and corrects any spelling mistakes in a text document. Designing a spell checker for Punjabi language is a challenging task. Punjabi language can be written in two scripts, Gurmukhi script (a Left to Right script based on Devanagari) and Perso-Arabic Script (a Right to Left script) which is also referred as Shahmukhi. Gurmukhi script follow ‘one sound - one symbol’ principle where as Shahmukhi follows ‘one sound - multiple symbol’ principle. Thus making Shahmukhi text even more challenging which complicates the design of spell checker for Shahmukhi text. The text written in Shahmukhi normally does not have short vowels and diacritic marks. So missing some of diacritic marks should not be considered as a mistake. But for Holy books like Quran, missing diacritic marks are considered as a mistake. So spell checker is designed in such a way that it can spell check with and without diacritic marks compulsion, which depends on user’s selection to spell check. In addition to this, Shahmukhi text has complex grammatical rules and phonetic properties. Thus it needs different algorithms and techniques for expected efficiency. This paper presents the complete design and implementation of a spell checker for Shahmukhi text.

Treasury of the True Dharma Eye: Zen Master Dogen’s Shobo Genzo ed. by Kazuaki Tanahashi (review)

Reviewed by: Treasury of the True Dharma Eye: Zen Master Dogen’sShobo Genzo ed. by Kazuaki Tanahashi Eitan Bolokan (bio) Treasury of the True Dharma Eye: Zen Master Dogen’s Shobo Genzo. Edited by Kazuaki Tanahashi. Boston: Shambhala, 2011. 2 vols. Pp. 1280. Hardcover $100.00, isbn978-1-590-30935-3. Kazuaki Tanahashi is one of the most influential translators of Zen Master Dōgen in recent decades. The scope and variety of his translations are as impressive as the manner in which he brings Zen, and especially Dōgen’s Zen, into modern English, serving as a rare bridge between cultures, ancient times, and modern society. Treasury of the True Dharma Eye: Zen Master Dogen’sShobo Genzo, a landmark translation of Dōgen’s magnum opus Shōbōgenzō, is the fourth project led by Tanahashi and sponsored by the San Francisco Zen Center, following Moon in a Dewdrop(New York: North Point, 1995), Enlightenment Unfolds(Boston: Shambhala, 2000), and Beyond Thinking(Boston: Shambhala, 2004), all of which provide selections from the Shōbōgenzōas well as other works by Dōgen. Treasury of the True Dharma Eyeconsists of two volumes. The first opens with important introductory remarks by the editor, and then presents the translations in chronological order, as dated by Dōgen, starting with his return from China in 1227 and up to the spring of 1245. The translations are divided into two groups. First is the “Wandering Period,” which includes the “Bendōwa,” and second is the “Kōshō Period,” which presents fascicles written between the “Makahannya-haramitsu” and “Kattō.” The second volume divides chronologically into four sections. “Monastery Construction Period” (1243–1245) presents fascicles between the “Sangai yuishin” and “Daishūgyō.” “Daibutsu Monastery Period” (1245–1246) presents the fascicles between “Kokū” and “Ōsaku sendaba.” “Eihei Period” (1246–1253) contains the fascicles between “Ji kuin mon” and “Hachidainin Gaku.” The last section consists of twelve undated fascicles. Concluding this volume are an Afterword by Michael Wenger, appendices, lineage charts, maps, a glossary, and a bibliography. The translations are based on Kōzen’s ninety-five-fascicle edition, published in the seventeenth century. In the section titled “Texts in Relation to Dōgen’s Life and Translation Credits” (vol. 1, pp. li–xcvii), Tanahashi clarifies that he and associate editor Peter Levitt used this edition as printed in Ōkubo Dōshū’s collection from 1970, the Dōgen Zenji zenshū(Complete works of Zen Master Dōgen), as well as the Dōgen Zenji zenshūversion edited by Sakai Tokugen et al. (vol. 1, p. li). Tanahashi and Levitt assert the need for various elucidations “to help readers decode the text” (vol. 1, p. xiv). Indeed, any reader of Dōgen will know how impenetrable and obscure [End Page 1286]his writings can be, highlighting just how valuable Tanahashi’s annotations are. Nonetheless, while applauding these efforts at making the texts more accessible, several editorial choices seem to merit further consideration—and certainly in light of the diverse audience drawn nowadays to Dōgen’s Zen. First, it is somewhat bewildering that a decision was made not to use macrons to indicate long vowels above Japanese terms, titles, and names, except in the glossary. Since Dōgen’s vocabulary is complex, these diacritical marks are important both technically for correct reading and culturally for the accurate transmission of Zen terminology in the West. Second, in both the table of contents and throughout the collection, fascicles are listed only in English, with no reference to their Japanese titles—even in transliteration. Considering that there are several extant English translations of the Shōbōgenzō, most of which disagree even on basic terms in Dōgen’s Zen, the Japanese originals may have served as a useful “anchor” for readers navigating between different editions. Furthermore, Japanese names and titles have their own vital role in reflecting Dōgen’s unique color of Zen, so I would hope to see them added to future editions of Treasury of the True Dharma Eye. A more complex issue concerns the decision to list fascicles in chronological order rather than along traditional editorial lines. As Tanahashi himself notes...

ZRCola vnašalni sistem za jezikoslovno rabo v programu word

Vnašalni sistem ZRCola je bil na Znanstvenoraziskovalnem centru SAZU v Ljubljani razvit za jezikoslovne, predvsem dialektološke potrebe in deluje v Microsoftovem programu word v operacijskem sistemu windows. Temelji na standardu unikod, okoli 1400 dodatnih znakov - med njimi so tudi slovenski zgodovinski znaki iz prve polovice 19. stoletja - pa je v pisavi 00 ZRCola razporejenih na področju zasebn e rabe tega standarda. Zapleteni znaki N se vnašajo tako, da se osnovnemu znaku v obveznem zaporedju dodajo zaslonski diakritični znaki, nazadnje pa program nize s temi znaki na uporabnikovo zahtevo sestavi v samostojne znake oz. pismenke. Ta način omogoča s prirejeno tipkovnico, kije del vnašalnega sistema, preprosto vnašanje latiničnih in slovanskih ciriličnih besedil in iskanje po njih, pa tudi lažjo izmenjavo datotek in zanesljivo organiziranje podatkovnih zbirk.

Many ways to read your vowels—Neural processing of diacritics and vowel letters in Hebrew

The current study examined the effect of orthographic transparency and familiarity on brain mechanisms involved in word recognition in adult Hebrew readers. We compared the effects of diacritics that provide transparent but less familiar information and vowel letters that increase orthographic transparency without compromising familiarity. Brain activation was measured in 18 adults during oral reading of single words, while manipulating the presence of diacritic marks, the presence of a vowel letter, and word length (3 vs. 4 consonants). We found opposite effects of diacritics and vowel letters on temporo-parietal regions associated with mapping orthography to phonology. The increase in activation for diacritic marks and the decrease in activation for vowel letters in these regions suggest that the greater familiarity of vowel letters compared to diacritics overrides the effect of orthographic transparency. Vowel letters also reduced activation in regions associated with semantic processing in unpointed words, and were thus distinct from the effect of an additional consonant. Altogether the results suggest that both orthographic transparency and familiarity contribute to word recognition.

Writing in Richardson variant of progressive supranuclear palsy in comparison to progressive non-fluent aphasia

BackgroundThe overlap between progressive supranuclear palsy (PSP) and progressive non-fluent aphasia (PNFA) is being increasingly recognized. In this paper descriptive writing in patients with Richardson syndrome of progressive supranuclear palsy (PSP-RS) is compared to writing samples from patients with PNFA. MethodsTwenty-seven patients participated in the study: 17 with the clinical diagnosis of PSP-RS and 10 with PNFA. Untimed written picture description was administered during neuropsychological assessment and subsequently scored by two raters blinded to the clinical diagnosis. Lexical and syntactic content, as well as writing errors (e.g. omission and perseverative errors) were analyzed. ResultsIn patients with PSP-RS both letter and diacritic mark omission errors were very frequent. Micrographia was present in 8 cases (47%) in PSP-RS group and in one case (10%) with PNFA. Perseverative errors did not differentiate between the groups. ConclusionsAs omission errors predominate in writing of patients with PSP-RS, writing seems to be compromised mainly because of oculomotor deficits, that may alter visual feedback while writing.

Text-to-speech synthesis system with Arabic diacritic recognition system

Text-to-speech synthesis system has been widely studied for many languages. However, speech synthesis for Arabic language has not sufficient progresses and it is still in its first stage. Statistical parametric synthesis based on hidden Markov models was the most commonly applied approach for Arabic language. Recently, synthesized speech quality based on deep neural networks was found as intelligible as human voice. This paper describes a Text-To-Speech (TTS) synthesis system for modern standard Arabic language based on statistical parametric approach and Mel-cepstral coefficients. Deep neural networks achieved state-of-the-art performance in a wide range of tasks, including speech synthesis. Our TTS system includes a diacritization system which is very important for Arabic TTS application. Our diacritization system is also based on deep neural networks. In addition to the use deep techniques, different methods were also proposed to model the acoustic parameters in order to address the problem of acoustic models accuracy. They are based on linguistic and acoustic characteristics (e.g. letter position based diacritization system, unit types based synthesis system, diacritic marks based synthesis system) and based on deep learning techniques (stacked generalization techniques). Experimental results show that our diacritization system can generate a diacritized text with high accuracy. As regards the speech synthesis system, the experimental results and subjective evaluation show that our proposed method for synthesis system can generate intelligible and natural speech.

Arabic character recognition using a Haar cascade classifier approach (HCC)

Optical character recognition (OCR) shows great potential for rapid data entry, but has limited success when applied to the Arabic language. Traditional OCR problems are compounded by the nature of Arabic language and because the script is heavily connected. A machine learning, Haar cascade classifier (HCC) approach was introduced by Viola and Jones (Rapid object detection using a boosted cascade of simple features. Kauai, Hawaii, 2001) to achieve rapid object detection based on a boosted cascade of simple Haar-like features. Here, that approach is applied for the first time to suit Arabic glyph recognition. HCC approach eliminates problematic steps in the pre-processing and recognition phases and, most importantly, character segmentation stage. A classifier was produced for each of the 61 Arabic glyphs that exist after the removal of diacritical marks (dots). These classifiers were trained and tested on some 2,000 images each. The system was tested with real text images and produces a recognition rate for Arabic glyphs of 87 %. The technique gives good results relative to those achieved using a commercial Arabic OCR application and existing state-of-the-art research application.

Spelling Error Patterns in Brazilian Portuguese

Fifty years after Damerau set up his statistics for the distribution of errors in typed texts, his findings are still used in a range of different languages. Because these statistics were derived from texts in English, the question of whether they actually apply to other languages has been raised. We address this issue through the analysis of a set of typed texts in Brazilian Portuguese, deriving statistics tailored to this language. Results show that diacritical marks play a major role, as indicated by the frequency of mistakes involving them, thereby rendering Damerau's original findings mostly unfit for spelling correction systems, although still holding them useful, should one set aside such marks. Furthermore, a comparison between these results and those published for Spanish show no statistically significant differences between both languages—an indication that the distribution of spelling errors depends on the adopted character set rather than the language itself.

Kalaupapa: A Collective Memory by Anwei Skinsnes Law (review)

Reviewed by: Kalaupapa: A Collective Memory by Anwei Skinsnes Law Colette J. Higgins Kalaupapa: A Collective Memory. By Anwei Skinsnes Law. Honolulu: University of Hawai‘i Press, 2012. xxi + 575 pp. Illustrated. Notes. Bibliography. Indexes. $28.99 paper; $45.00 cloth There have been many books written about Kalaupapa as a place where those with Hansen’s Disease (a.k.a. leprosy) were isolated. Previous books have typically focused on Kalaupapa’s notable mea kōkua (helpers), such as Father. Damien, Mother Marianne, and Brother Dutton, but Kalauapapa: A Collective Memory by Anwei Skinsnes Law is unique in that it tells the story of this place from the perspective of those being sent there by the government’s isolation policy. Law’s objective in writing this book is to bring back the voices of the people themselves. She is able to do this by using extensive quotes discovered after decades of archival research and after collecting over 200 hours of oral history interviews. For anyone interested in leprosy, Kalaupapa, or Hawaiian history, this is a must read. This book is appropriately subtitled A Collective Memory. While its chronological organization and easy-to-read prose make it an ideal textbook for a college course about Kalaupapa or leprosy in Hawai‘i, it does not read like a typical textbook. It is in fact a collection of many memories over many years. More than 300 photographs and long excerpts from primary sources allows readers to immerse themselves into an experience that feels like it is part photo album and part memoir. Additionally, this book contains seventeen pages of endnotes and a comprehensive bibliography that will assist future researchers who want to dig deeper into specific events or people. For those doing genealogical research about family members who were sent to Kalaupapa, the Index of Names will serve as a useful tool. Concerning the Hawaiian language style, the author notes that diacritical marks, i.e. the ‘okina (glottal) and kahakō (macron), are “not generally used in this book” because the names and quotes are “primarily from the time period before these symbols were used” (p. xxi). Considering that over 8,000 people were sent to Kalaupapa between 1866 and 1969, this book attempts to resurrect as many stories of individuals as possible. To connect these individual stories, Law provides some consistent themes throughout her book. She highlights their individual identities by focusing on their unique abilities. This is possible because she has had more than 300 pages of letters and documents translated from Hawaiian to English. The letters to the Board of Health and the Hawaiian language newspapers show a prolonged nonviolent resistance by the people themselves. They refused to accept the unjust isolation policy that separated them from their families, they insisted on their rights as human beings and to be treated [End Page 204] with dignity, and their writings reveal no shame or stigma associated with leprosy within the Hawaiian community. Law provides a very clear contrast of the two cultural responses to leprosy (foreign vs. Hawaiian). The Board of Health represents the foreign view that sought to isolate those with the disease, while the Hawaiian cultural response was to care for their sick family members. The letters reveal numerous requests to the Board of Health seeking permission to serve as mea kōkua for their sick relatives sent to Kalaupapa. Law’s use of so many primary sources reveals a clear contrast in cultural responses when comparing English vs. Hawaiian sources. This book includes chapters on Father Damien De Veuster, Mother Marianne Cope, and Joseph Dutton, but from the perspective of the residents they served. Also included are the early leaders in the Kalaupapa community (i.e. William Humphreys, Jonathan Napela, Peter Kaeo, Ambrose Hutchison, Thomas Nathaniel). Of these, Ambrose Hutchison deserves a special mention. He lived in Kalauapapa from 1879 to 1932, served in an official leadership capacity for ten years, and has largely been left out of Kalaupapa’s history, until now, thanks to Law’s extensive use of his unpublished manuscript. A student looking for a research paper topic might want to dig deeper into one of the following: the inoculation of Keanu and whether he could...