54,841 Hits in 3.8 sec

A General Language Assistant as a Laboratory for Alignment [article]

Amanda Askell, Yuntao Bai, Anna Chen, Dawn Drain, Deep Ganguli, Tom Henighan, Andy Jones, Nicholas Joseph, Ben Mann, Nova DasSarma, Nelson Elhage, Zac Hatfield-Dodds (+10 others)
2021 arXiv   pre-print
Given the broad capabilities of large language models, it should be possible to work towards a general-purpose, text-based assistant that is aligned with human values, meaning that it is helpful, honest  ...  We find that the benefits from modest interventions increase with model size, generalize to a variety of alignment evaluations, and do not compromise the performance of large models.  ...  We thank Paul Christiano for helpful discussions.  ... 
arXiv:2112.00861v3 fatcat:g7awtuczp5aldjdl75rwf2eaxm

Improving End-to-end Speech Recognition with Pronunciation-assisted Sub-word Modeling [article]

Hainan Xu, Shuoyang Ding, Shinji Watanabe
2019 arXiv   pre-print
Most end-to-end speech recognition systems model text directly as a sequence of characters or sub-words.  ...  We propose pronunciation-assisted sub-word modeling (PASM), a sub-word extraction method that leverages the pronunciation information of a word.  ...  Extraction As fast align is a re-parameterization of IBM model 2, a typical alignment method for statistical machine translation, it does not limit itself in generating Figure 2 , as well as "null-alignments  ... 
arXiv:1811.04284v2 fatcat:s4s5lxwmcrfzxm5vt65u6ggg7a

Translation ambiguity in and out of context

2010 Applied Psycholinguistics  
These findings establish the utility of parallel language corpora as important tools in psycholinguistic investigations of bilingual language processing.  ...  We compare translations of single words, made by bilingual speakers in a laboratory setting, with contextualized translation choices of the same items, made by professional translators and extracted from  ...  However, it is not likely that such occasions would outnumber the cases where translation from a third source language would result in less alignment between the corpus and the laboratory data.  ... 
doi:10.1017/s0142716410000305 fatcat:yajygeu4xzgmpgsyiomt3wo4tu

Multimodal Machine Learning: Integrating Language, Vision and Speech

Louis-Philippe Morency, Tadas Baltrušaitis
2017 Proceedings of ACL 2017, Tutorial Abstracts  
With the initial research on audio-visual speech recognition and more recently with language & vision projects such as image and video captioning and visual question answering, this research field brings  ...  some unique challenges for multimodal researchers given the heterogeneity of the data and the contingency often found between modalities.  ...  For example, language is often seen as symbolic while audio and visual modalities will be represented as signals. 2.  ... 
doi:10.18653/v1/p17-5002 dblp:conf/acl/MorencyB17 fatcat:m24h75t6mvdyfeedrsjbvjjaom

Arabic Language Learning Assisted by Computer, based on Automatic Speech Recognition [article]

Naim Terbeh, Mounir Zrigui
2012 arXiv   pre-print
This work consists of creating a system of the Computer Assisted Language Learning (CALL) based on a system of Automatic Speech Recognition (ASR) for the Arabic language using the tool CMU Sphinx3 [1],  ...  To this work, we have constructed a corpus of six hours of speech recordings with a number of nine speakers. we find in the robustness to noise a grounds for the choice of the HMM approach [2]. the results  ...  in the laboratory of Grenoble.  ... 
arXiv:1205.3316v1 fatcat:2pvicuutvff67f6ojuuyxm37mq

Participatory Philology: Computational Linguistics and the Future of Historical Language Education

Gregory Crane, Stella Rose Dee, Anna Krohn
2014 Human Computation  
The paper presents the work of the Historical Languages eLearning Project and the incorporation of pedagogy into resources for participatory philology and reviews a case study of a class at Tufts University  ...  With a focus on the interface between 21st-century philology and citizen science, this paper reviews the data we collect, why we gather that data, and the cohort that we engage for assistance with data  ...  Annotation for Research 3.1 Translation Alignment Translation alignment describes the process of annotating the correspondences between the words of a modern language translation with a historical language  ... 
doi:10.15346/hc.v1i2.8 dblp:journals/hc/CraneDK14 fatcat:jbev5klv45efda4lpy5orivn24

Statistical phone duration modeling to filter for intact utterances in a computer-assisted pronunciation training system

Wai-Kit Lo, Alissa M. Harrison, Helen Meng
2010 2010 IEEE International Conference on Acoustics, Speech and Signal Processing  
We study the use of a statistical phone duration model for separating intact utterances from corrupted ones in a computer-assisted pronunciation training system.  ...  For these utterances, our system performs detailed phonetic analysis of the alignment and generates corrective feedback to highlight the occurrence of phonetic errors.  ...  Phone duration has been used as feature for computing confidence measures in ASR applications for embedded and noise environments [7, 8] as well as verifying selected utterances in a language learning  ... 
doi:10.1109/icassp.2010.5494988 dblp:conf/icassp/LoHM10 fatcat:mvy6x6poenekdgkt4qxjg4jy4u

Developing of guidance for laboratory practice of islamic science-integrated plant anatomy-physiology

Ahmad Ali, Wiwin Pramita Arif
2019 Biosfer  
The research aims to develop a valid, effective, and practice guidance for laboratory practice of Islamic science-integrated plant anatomy-physiology.  ...  Effective criteria with 92% of students who performed the laboratory practice give positive responses to the overall item questioned, and 100% assistants provide positive response on 93.33% items asked  ...  The guidance is used as a guide in laboratory practice work steps for students as well as assistants/lecturers.  ... 
doi:10.21009/biosferjpb.v12n1.70-82 fatcat:s7trwq6rpbaffhghnuiez2i77y

An Ontology-Based Method for Semantic Integration of Business Components

Hicham Elasri, Abderrahim Sekkaki, Larbi Kzaz
2011 2011 11th Annual International Conference on New Technologies of Distributed Systems  
Our contribution concerns both the proposal of a BC integration solution based on ontologies alignment and a method for enriching the domain ontology used as a support for alignment.  ...  This work is positioned at the intersection of two research areas: Integration of reusable Business Components and alignment of ontologies for semantic conflict resolution.  ...  Business Components to integrate will be used as a basis for the generation of new semantic relations.  ... 
doi:10.1109/notere.2011.5957993 dblp:conf/notere/ElasriSK11 fatcat:whg7pans2jgzfonsvsw6g66gs4

Statistical Sign Language Machine Translation: from English written text to American Sign Language Gloss [article]

Achraf Othman, Mohamed Jemni
2011 arXiv   pre-print
This works aims to design a statistical machine translation from English text to American Sign Language (ASL).  ...  The system is based on Moses tool with some modifications and the results are synthesized through a 3D avatar for interpretation. First, we translate the input text to gloss, a written form of ASL.  ...  In the 18th century, deaf people are beginning the use of a Sign Language (SL) based on gestural actions. Gestures that can express human thought as much as a spoken language.  ... 
arXiv:1112.0168v1 fatcat:y4m4khcizfcsvlvj2qjumte6im

A Hybrid Model for Enhancing Lexical Statistical Machine Translation (SMT) [article]

Ahmed G. M. ElSayed, Ahmed S. Salama, Alaa El-Din M. El-Ghazali
2015 arXiv   pre-print
incorporating different statistical and Natural Language Processing (NLP) models such as language model, alignment model, phrase based model, reordering model, and translation model.  ...  A proposed Statistical Machine Translation (SMT) based model that can be used to translate a sentence from the source Language (English) to the target language (Arabic) automatically through efficiently  ...  Clustering as a general approach for dealing with issues of language sparsity and morphological analysis is promising.  Use large amount of parallel corpus data to train the proposed model, to achieve  ... 
arXiv:1506.01171v1 fatcat:gmognnfddbfsjgoo3yinzyxele

Mapping Unparalleled Clinical Professional and Consumer Languages with Embedding Alignment [article]

Wei-Hung Weng, Peter Szolovits
2018 arXiv   pre-print
In this work, we utilized the embeddings alignment method for the word mapping between unparalleled clinical professional and consumer language embeddings.  ...  We show that the Procrustes algorithm can be performant for the professional consumer language embeddings alignment, whereas adversarial training with refinement may find some relations between two languages  ...  Therefore, how to translate clinical professional language to consumer-level language is essential to improve the communication between consumers and clinicians, as well as to assist consumers' decision  ... 
arXiv:1806.09542v1 fatcat:7ggvnxu6zvevrnkvprauucr2wa

Molecules to maps: tools for visualization and interaction in support of computational biology

E. T. Kraemer, T. E. Ferrin
1998 Bioinformatics  
The volume of data produced by genome projects, X-ray crystallography, NMR spectroscopy, and electron and confocal microscopy present the bioinformatics community with new challenges for analyzing, understanding  ...  applications running across multiple platforms, and a 'virtual laboratory notebook' for managing experiments with simulation models.  ...  The sequence alignment editor presents multiple protein sequences. Each sequence is represented as a series of residue tiles.  ... 
doi:10.1093/bioinformatics/14.9.764 pmid:9918946 fatcat:x4feoxu7xzdafjl4a32qmlmn74

Crowdsourcing the Paldaruo Speech Corpus of Welsh for Speech Technology

Cooper, Jones, Prys
2019 Information  
Collecting speech data for a low-resource language is challenging when funding and resources are limited.  ...  These methods can be extended to other languages, and suggestions for other low-resource languages are discussed.  ...  Acknowledgments: We thank David Chan, research officer on the original GALLU project for assistance with data design, collection and software.  ... 
doi:10.3390/info10080247 fatcat:zfrxmly5yrep7gjcdggyz7odai

Intelligent Links: AI-Supported Connections between Employers and Colleges

Robby Robson, Elaine Kelsey, Ashok Goel, Sazzad Nasir, Elliot Robson, Myk Garn, Matt Lisle, Jeanne Kitchens, Spencer Rugaber, Fritz Ray
2022 The AI Magazine  
When modernization and other changes demand workforce reskilling, employers often turn to local colleges for training programs. Doing so can be a frustrating experience.  ...  taxonomies, uses machine-learned models to align skills needs with learning outcomes and training, and incorporates an intelligent coach based on Georgia Tech's Jill Watson "virtual teaching assistant  ...  However, for KSA extraction and alignment of training requests with course content and providers, we pretrained a BERT-based general language model from scratch.  ... 
doi:10.1609/aimag.v43i1.19128 fatcat:s6mrdxkg5nddrarhzppnkfzaei
« Previous Showing results 1 — 15 out of 54,841 results