Academia.edu no longer supports Internet Explorer.
To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to upgrade your browser.
2023, Cambridge Language Sciences Annual Symposium
By the end of the century, over half of the 6500 languages spoken in the world will die out (Turin, 2007). Nepal's situation is particularly dire: of the 120+ distinct languages identified in the 2011 census, 60 are endangered due to globalisation, socio-political unrest, and environmental challenges. The loss of these languages also means the loss of unique cultural and religious identifiers. Given this, there is a need for methods and tools to preserve linguistic diversity. A major challenge in language preservation, however, is the transcription bottleneck (Shi et al., 2021): transcribing one minute of audio requires an average of 40+ minutes (Durantin et al., 2017). This becomes even more complicated for endangered languages with no (standardised) orthographies or documentation. While advanced automatic speech-recognition (ASR) tools are available, they are often ineffective for these extremely low-resource languages (Foley et al., 2018). This poster presents the preliminary results to address these issues for the Newar and Dzardzongke (both representing different branches of the Sino-Tibetan language family, spoken in Nepal) using Wav2Vec2 models fine-tuned for low-resource languages (Coto-Solano 2021, 2022). We show that endangered languages benefit from a specific set of optimisation procedures through tests with Kaldi vs Wav2Vec2; different types of data augmentation, and the development of a new or standardisation of orthography.
2021 •
There is growing interest in ASR systems that can recognize phones in a language-independent fashion. There is additionally interest in building language technologies for low-resource and endangered languages. However, there is a paucity of realistic data that can be used to test such systems and technologies. This paper presents a publicly available, phonetically transcribed corpus of 2255 utterances (words and short phrases) in the endangered Tangkhulic language East Tusom (no ISO 639-3 code), a Tibeto-Burman language variety spoken mostly in India. Because the dataset is transcribed in terms of phones, rather than phonemes, it is a better match for universal phone recognition systems than many larger (phonemically transcribed) datasets. This paper describes the dataset and the methodology used to produce it. It further presents basic benchmarks of state-of-the-art universal phone recognition systems on the dataset as baselines for future experiments.
arXiv (Cornell University)
Leveraging End-to-End ASR for Endangered Language Documentation: An Empirical Study on Yolox\'ochitl Mixtec2021 •
Text, Speech and …
Towards Automatic Transcription of Large Spoken Archives in Agglutinating LanguagesHungarian ASR for the MALACH Project2007 •
Transcribing speech is an important part of language documentation, yet speech recognition technology has not been widely harnessed to aid linguists. We explore the use of a neural network architecture with the connectionist temporal classification loss function for phonemic and tonal transcription in a language documentation setting. In this framework, we explore jointly modelling phonemes and tones versus modelling them separately, and assess the importance of pitch information versus phonemic context for tonal prediction. Experiments on two tonal languages, Yongning Na and Eastern Chatino, show the changes in recognition performance as training data is scaled from 10 minutes up to 50 minutes for Chatino, and up to 224 minutes for Na. We discuss the findings from incorporating this technology into the linguistic workflow for documenting Yongning Na, which show the method's promise in improving efficiency, minimizing typographical errors, and maintaining the transcription's faithfulness to the acoustic signal, while highlighting phonetic and phonemic facts for linguistic consideration.
Proceedings of the 6th Intl. Workshop on Spoken Language Technologies for Under-Resourced Languages
Building Speech Recognition Systems for Language Documentation: The CoEDL Endangered Language Pipeline and Inference System (ELPIS)2018 •
Machine learning has revolutionised speech technologies for major world languages, but these technologies have generally not been available for the roughly 4,000 languages with populations of fewer than 10,000 speakers. This paper describes the development of ELPIS, a pipeline which language documentation workers with minimal computational experience can use to build their own speech recognition models, resulting in models being built for 16 languages from the Asia-Pacific region. ELPIS puts machine learning speech technologies within reach of people working with languages with scarce data, in a scalable way. This is impactful since it enables language communities to cross the digital divide, and speeds up language documentation. Complete automation of the process is not feasible for languages with small quantities of data and potentially large vocabularies. Hence our goal is not full automation, but rather to make a practical and effective workflow that integrates machine learning technologies.
2016 •
Conference of the International Speech Communication Association
GlobalTIMIT: Acoustic-Phonetic Datasets for the World's Languages2018 •
2023 •
Slides for a paper presented at the 2023 symposium "Tikal and its Wooden Memory" at the Museum Der Kulturen, Basel, Switzerland. October 13, 2023. This talk looks at the narrative structures of the wooden lintels in two of Tikal's major temples, proposing that they were placed in their original setting above or in proximity to the captured effigies of foreign deities (Yamanajaw and Waxaklajuun Ubaah Chan in Temple I, and the patron deities of El Peru and Naranjo in Temple IV). While the physical effigies were likely place within the small confines of the structures, the narratives also emphasize the Tikal rulers' ritual ability to embody or "impersonate" these deities. I argue that the lofty temples were settings for the display and presentation of the deities and/or the victorious deified rulers, and Temple IV at least may have been intentionally built too be oriented directly toward Tikal's enemy, Naranjo, to the east.
Pensar rapido, pensar despacio
Pensar rapido, pensar despacio Daniel Kahneman2012 •
Daniel Kahneman, uno de los pensadores más importantes del mundo, recibió el premio Nobel de Economía por su trabajo pionero en psicología sobre el modelo racional de la toma de decisiones. Sus ideas han tenido un profundo impacto en campos tan diversos como la economía, la medicina o la política, pero hasta ahora no había reunido la obra de su vida en un libro. En Pensar rápido, pensar despacio, un éxito internacional, Kahneman nos ofrece una revolucionaria perspectiva del cerebro y explica los dos sistemas que modelan cómo pensamos. El sistema 1 es rápido, intuitivo y emocional, mientras que el sistema 2 es más lento, deliberativo y lógico. Kahneman expone la extraordinaria capacidad (y también los errores y los sesgos) del pensamiento rápido, y revela la duradera influencia de las impresiones intuitivas sobre nuestro pensamiento y nuestra conducta. El impacto de la aversión a la pérdida y el exceso de confianza en las estrategias empresariales, la dificultad de predecir lo que nos hará felices en el futuro, el reto de enmarcar adecuadamente los riesgos en el trabajo y en el hogar, el profundo efecto de los sesgos cognitivos sobre todo lo que hacemos, desde jugar en la Bolsa hasta planificar las vacaciones; todo esto solo puede ser comprendido si entendemos el funcionamiento conjunto de los dos sistemas a la hora de formular nuestros juicios y decisiones. Al implicar al lector en una animada reflexión sobre cómo pensamos, Kahneman consigue revelar cuándo podemos confiar en nuestras intuiciones y cuándo no, y de qué modo podemos aprovechar los beneficios del pensamiento lento. Además, ofrece enseñanzas prácticas e iluminadoras sobre cómo se adoptan decisiones en la vida profesional o personal, y sobre cómo podemos usar distintas técnicas para protegernos de los fallos mentales que nos crean problemas. Pensar rápido, pensar despacio cambiará para siempre nuestra manera de pensar sobre cómo pensamos. www.lectulandia.com-Página 2
Quarterly Journal of Management and Development Process
موانع دستیابی به اجرای کامل خطمشیهای عمومی1999 •
University of Minnesota Press
Empirical Ecocriticism: Environmental Narratives for Social Change2023 •
Brazilian Journal of Infectious Diseases
Liver cirrhosis and rhino-orbital mucormycosis, a possible but rare association: description of a clinical case and literature review2009 •
Aquatic Conservation: Marine and Freshwater Ecosystems
Distribution and status of medicinal leeches (genus Hirudo) in the Western Palaearctic: anthropogenic, ecological, or historical effects?2010 •
Majalah Kedokteran Gigi Indonesia
Evaluasi Hasil Preparasi Servikal pada Model Kerja Gigi Tiruan Jembatan2014 •
CorSalud (Revista de Enfermedades Cardiovasculares)
P-wave dispersion in hypertensive children; its relationship with body weight2012 •
Jurnal Ilmiah Teknologi Infomasi Terapan
Aplikasi "Rekam Medis" Menggunakan Metode Prototyping Pada UPT.Puskesmas Kalitanjung Cirebon2021 •