Morphological decomposition in Arabic ASR systems

In recent years, the use of morphological decomposition strategies for Arabic Automatic Speech Recognition (ASR) has become increasingly popular. Systems trained on morphologically decomposed data are often used in combination with standard word-based approaches, and they have been found to yield consistent performance improvements. The present article contributes to this ongoing research endeavour by exploring the use of the 'Morphological Analysis and Disambiguation for Arabic' (MADA) tools for this purpose. System integration issues concerning language modelling and dictionary construction, as well as the estimation of pronunciation probabilities, are discussed. In particular, a novel solution for morpheme-to-word conversion is presented which makes use of an N-gram Statistical Machine Translation (SMT) approach. System performance is investigated within a multi-pass adaptation/combination framework. All the systems described in this paper are evaluated on an Arabic large vocabulary speech recognition task which includes both Broadcast News and Broadcast Conversation test data. It is shown that the use of MADA-based systems, in combination with word-based systems, can reduce the Word Error Rates by up to 8.1 relative. © 2012 Elsevier Ltd. All rights reserved.

Graphone model interpolation and Arabic pronunciation generation

This paper extends n-gram graphone model pronunciation generation to use a mixture of such models. This technique is useful when pronunciation data is for a specific variant (or set of variants) of a language, such as for a dialect, and only a small amount of pronunciation dictionary training data for that specific variant is available. The performance of the interpolated n-gram graphone model is evaluated on Arabic phonetic pronunciation generation for words that can't be handled by the Buckwalter Morphological Analyser. The pronunciations produced are also used to train an Arabic broadcast audio speech recognition system. In both cases the interpolated graphone model leads to improved performance. Copyright © 2011 ISCA.

Word boundary modelling and full covariance gaussians for Arabic Speech-to-Text systems

This paper describes recent improvements to the Cambridge Arabic Large Vocabulary Continuous Speech Recognition (LVCSR) Speech-to-Text (STT) system. It is shown that wordboundary context markers provide a powerful method to enhance graphemic systems by implicit phonetic information, improving the modelling capability of graphemic systems. In addition, a robust technique for full covariance Gaussian modelling in the Minimum Phone Error (MPE) training framework is introduced. This reduces the full covariance training to a diagonal covariance training problem, thereby solving related robustness problems. The full system results show that the combined use of these and other techniques within a multi-branch combination framework reduces the Word Error Rate (WER) of the complete system by up to 5.9% relative. Copyright © 2011 ISCA.

Catalogue of Irish Manuscripts in Villanova University, Pennsylvania

RAE2008

"A Good Reder": the middle English Wise Book of Philosophy and Astronomy, instruction, publics, and manuscripts

The present work is a study of the Middle English prose text known as The Wise Book of Philosophy and Astronomy, a consideration of its transmission and reception history, and a survey of its manuscript witnesses; it also incorporates an edition of the text from two of its manuscripts. The text is a cosmological treatise of approximately five thousand words, written for the most part in English, with astronomical and astrological terms in Latin, though the English translation is frequently given. It is written anonymously, and survives in thirty-three manuscripts.

The Oud Across Arabic Culture

This study is a compilation and compendium of information on the oud, the most important instrument in Arabic classical music. It has grown out of my own long-time involvement in studying and playing the oud, and in particular out of my interest in the lack of sources and knowledge available to the vast majority of oud players and researchers, as well as for the readers. My own path started from an intensive study of the oud, which included exposure to several treaties; some housed in museums around the globe, and some only available in the Arabic language. The study combines archival research (including Arabic poetry and pre-Islamic Era and medieval treaties), symbolism, new archaeological discoveries, field interviews, and analysis of existing scholarship, and draws on my professional performance experience for detailed stylistic analysis of the oud's performance practice and its historical development. The study consists of participant observation, personal performance, and interviews conducted in person, via telephone, and/or via e-mail, according to the choice of the performers. The performers have been selected from networks of musicians who perform regularly at lounges, concert halls, and private events. These performers have been chosen according to their musical knowledge, technical skill, experience, and activity in Arabic music and oud performance. Chapter one deals with the purpose of this study and the methods of investigation, as well as giving a brief overview of the history of the oud. In addition, there will be an introduction to the Arabic musical system (mâqâm), which is primarily based on the mechanics and sound production of the oud. Chapter two deals with the oud in Arabic sources: the first source is Arabic poetry in the pre-Islamic Era. The second source is Arabic poetry in the medieval era, in which I found a significant number of poets who allude to the oud, providing accurate descriptions of the player, singers, and the scenes within the contexts of oud performance. The third source is the Arab scholars' intensive treatises with meticulous accounts of the instrument's apparatii, including descriptions and measurements of the parts, strings, and tuning. While chapter three deals with the classification, the development of the oud, chapter four deals with topics such as: the symbolism of the oud and its relation to cosmology, astronomy, mathematics and anatomy. In most of the pertinent Arabic writings, philosophers mention a significant correlation between the oud and the other sciences. Chapter five deals with recreating the performance practice of the oud. A case study of the oud performers focuses on their style, technique, training, and personal experiences. Topics such as improvisation and ornamentation, the oud in the Arabic musical ensemble, the social uses and functions, and gender in musical performance practices will be included in detailed analysis. Other important topics will be analyzed such as traditional vs. modern technique, and the repertoire of the oud. Specifically, in regard to technique, the study outline the style of the music, the role of the oud in Arabic ensembles, the function of the oud in music composition, and the form of the ensembles in Arabic performance and practice.

Hope in the Most Unlikely Spaces: Thawra and the Contemporary Arabic Novel

In the early to mid-twentieth century, many novelists in the Arab world championed Arab nationalism in their literary reflections on the social and political struggles of their countries, depicting these struggles primarily in terms of spatial binaries that pitted the Arab world against the West, even as they imported Western literary models of progress and modernity into their own work. The intense experience of national awakening that infused their writing often placed these authors at a literary disadvantage, for in their literature, all too often the depth and diversity of Arabic cultures and the complexity of socio-political struggles across the Arab world were undermined by restrictive spatial discourses that tended to focus only on particular versions of Arab history and on a seemingly unifying national predicament. Between the Arab defeat of 1967 and the present day, however, an increasing number of Arab authors have turned to less restrictive forms of spatial discourse in search of a language that might offer alternative narratives of hope beyond the predictable, and seemingly thwarted, trajectories of nationalism. This study traces the ways in which contemporary Arab authors from Egypt and the Sudan have endeavoured to re-think and re-define the Arab identity in ever-changing spaces where elements of the local and the global, the traditional and the modern, interact both competitively and harmoniously. I examine the spatial language and the tropes used in three Arabic novels, viewing them through the lens of thawra (revolution) in both its socio-political and artistic manifestations. Linking the manifestations of thawra in each text to different scenes of revolution in the Arab world today, in Chapter Two, I consider how, at a stage when the Sudan of the sixties was both still dealing with colonial withdrawal and struggling to establish itself as a nation-state, the geographical and textual landscapes of Tayeb Salih‟s Season of Migration to the North depict the ongoing dilemma of the Sudanese identity. In Chapter Three, I examine Alaa iii al-Aswany‟s The Yacoubian Building in the context of a socially diseased and politically corrupt Egypt of the nineties: social, political, modern, historical, local, and global elements intertwine in a dizzyingly complex spatial network of associations that sheds light on the complicated reasons behind today‟s Egyptian thawra. In Chapter Four, the final chapter, Gamal al-Ghitani‟s approach to his Egypt in Pyramid Texts drifts far away from Salih‟s anguished Sudan and al-Aswany‟s chaotic Cairo to a realm where thawra manifests itself artistically in a sophisticated spatial language that challenges all forms of spatial hegemony and, consequently, old and new forms of social, political, and cultural oppression in the Arab world.