Keynote4 – Machine Translation at the Government of Canada – Reaching for the Future
G1 -Neural Translator Designed to Protect the Eastern Border of the European Union G2 – Dragonfly – Improving Automated Sign Language Recognition (ASLR) with Synthetic Data
G3 – Using OpenNMT on Android devices with TensorFlow Lite G4 – Corpus Creation and Evaluation for Speech-to-Text and Speech Translation
R1 – Learning Curricula for Multilingual Neural Machine Translation Training R2 – Investigating Active Learning in Interactive Neural Machine Translation
R3 – Crosslingual Embeddings are Essential in UNMT for distant languages R4 – Neural Machine Translation in Low-Resource Setting – a Case Study in English-Marathi Pair R5 – Transformers for Low-Resource Languages – Is Féidir Linn! R6 – The Effect of Domain and Diacritics in Yoruba–English Neural Machine Translation R7 – Integrating Unsupervised Data Generation into Self-Supervised NMT for Low-Resource Languages
R8 – Surprise Language Challenge – Developing NMT between Pashto and English in Two Months R9 – Like Chalk and Cheese? On the Effects of Translationese in MT Training
R10 – Investigating Softmax Tempering for Training Neural Machine Translation Models R11 – Scrambled Translation Problem – A Problem of Denoising UNMT R12 – Make the Blind Translator See The World – A Novel Transfer Learning Solution for Multimodal MT R13 – Sentiment Preservation in Review Translation using Curriculum-based Re-inforcement Framework R14 – On nature and causes of observed MT errors
R15 – A Comparison of Sentence-Weighting Techniques for NMT R16 – Sentiment-based Candidate Selection for NMT
R17 – Studying The Impact Of Document-level Context On Simultaneous Neural Machine Translation R18 – Attainable Text-to-Text MT vs. Translation – Issues Beyond Linguistic Processing R19 – Modeling Target-side Inflection in Placeholder Translation
R20 – Product Review Translation using Phrase Replacement and Attention Guided Noise Augmentation R21 – Optimizing Word Alignments with Better Subword Tokenization
R22 – Introducing Mouse Actions into Interactive-Predictive Neural Machine Translation R23 – Neural Machine Translation with Inflected Lexicon R24 – An Alignment-Based Approach to Semi-Supervised Bilingual Lexicon Induction
Tutorial 1 – Introduction to Machine Translation
Tutorial 2 – A Deep Learning Curve for Post-Editing
Tutorial 3 – Training an industry grade NMT model – Tips & Tricks
Tutorial 4 – Theory and Practice for research in Post-editese
UP3 – Selecting the best data filtering method for NMT training UP4 – A Review for Large Volumes of Post-edited Data UP5 – Accelerated Human NMT Evaluation Approaches for NMT Workflow Integration UP6 – MT Human Evaluation – Insights & Approaches UP7 – A Rising Tide Lifts All Boats? Quality Correlation of Human and Machine Assisted Translation
UP8 – Bad to the Bone – Predicting the Impact of Source on MT UP9 – MTPE from the Perspective of Translation Trainees – Implications for Translation Pedagogy
UP10 – Using Raw MT to make essential information available for a range of potential customers UP11 – Field Experiments of Real Time Foreign News Distribution Powered by MT UP12 – Data-Driven MT Adoption UP13 – Preserving high MT quality for content with inline tags UP14 – Early-stage development of the SignON app and open framework – challenges and opportunities
UP15 – Deploying MT Quality Estimation on a large scale – Lessons learned and open questions UP16 – Validating Quality Estimation in a CAT Workflow – Speed, Cost and Quality Trade-off
UP17 – Neural Translation for European Union (NTEU) UP18 – A Data-Centric Approach to Real-World Custom NMT for Arabic UP19 – Building MT systems for Public Sector users in Croatia, Iceland, Ireland, and Norway
UP20 – Using speech technology in the translation process workflow in international organizations UP21 – Multi-Domain Adaptation in Neural Machine Translation Through Multidimensional Tagging
UP22 – cushLEPOR uses LABSE distilled knowledge to improve correlation with human translations UP23 – A Synthesis of Human and Machine – Correlating “New” Auto-Eval Metrics with Human Assessments UP24 – Lab vs. Production – Two Approaches to Productivity Evaluation for MTPE for LSP
Workshop 1 – The 9th Workshop on Patent and Scientific Literature Translation (PSLT2021)
Workshop 2 – The 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021)
Workshop 2 – The 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021) – Part 2
Workshop 2 – The 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021) – Part 3
Workshop 3 – Automated Spoken Language Translation in Real-World Settings (ASLTRW)
Workshop 3 – Automated Spoken Language Translation in Real-World Settings (ASLTRW) – Part 2
Workshop 4 – 1st International Workshop on Automatic Translation for Signed and Spoken Languages
Workshop 4 – 1st International Workshop on Automatic Translation for Signed and Spoken Languages – Part 2
Roundtable – Building MT Capacity and Competence In-House
C3 – A Survey of Qualitative Error Analysis for Neural Machine Translation Systems
https://youtu.be/0kF57rJFiEo
C10 – Flexible Customization of a Single Neural MT System with Multi dimensional Metadata Inputs
https://youtu.be/NzxIn_YbGK4
K1 – Welcome and Keynote – MT Developments in the European Union
https://youtu.be/0o6R0gRsJ5g
K2 – Faithfulness in natural language generation in an era of heightened ethical AI awareness
https://youtu.be/obewkNfXZLw
K3 – Welcome and Keynote – Factor 1000 Better MT, more content, and what we can do with it
https://youtu.be/K5CMWXIdywk
K4 – Keynote – Navigating Change Keys to implementing language technology in government
https://youtu.be/b0n3p_ODAdc
K5 – Using language technology to enable two way communication in humanitarian assistance
https://youtu.be/t_aU3AWwsx4
K6 – Keynote – Research stories from Google Translate’s Transcribe Mode
https://youtu.be/9hSX7DUv240
G1 – Successful Tech Transfer of MT Research in Government
https://youtu.be/TBJVVpCIzJY
G2 – Plugging into Trados Augmenting Translation in the Enclave
https://youtu.be/Nk4D7Zqe4cU
G3 – PEMT for the Public Sector Discovery, Scoping, and Delivery
https://youtu.be/OHNlVOnUtrI
G4 – Shareable TTS Components
https://youtu.be/LQGNVirs4vQ
G5 – A Tale of Eight Countries or the EU Council Presidency Translator in Retrospect
https://youtu.be/g0df3qibbAw
G6 – American Sign Language ASL to English Machine Translation
https://youtu.be/cKCKTuGpyLE
G7 – Why is it so Hard to Develop Comparable Translation Evaluations and How Can Standards Help
https://youtu.be/-b5dRXYs9-U
G8 – Using Contemporary US Government Data to Train Custom MT for COVID 19
https://youtu.be/EL3ABezbC_c
R3 – Machine Translation with Unsupervised Length Constraints
https://youtu.be/l77muNlzYaE
R4 – Constraining the Transformer NMT Model with Heuristic Grid Beam Search
https://youtu.be/TtdlO3GEl9o
R6 – Generative latent neural models for automatic word alignment
https://youtu.be/p0SwNFHHmCg
R7 – The Impact of Indirect Machine Translation on Sentiment Classification
https://youtu.be/lvNMT6jG-AY
R8 – Towards Handling Compositionality in Low Resource Bilingual Word Induction
https://youtu.be/g57HdwlVBLI
R9 – The OpenNMT Neural Machine Translation Toolkit 2020 Edition
https://youtu.be/2_VMqRkT1zg
R10 – The Sockeye 2 Neural Machine Translation Toolkit at AMTA 2020
https://youtu.be/QnR3kZ3oD1Y
R11 – THUMT An Open Source Toolkit for Neural Machine Translation
https://youtu.be/xSsE8NNoaoY
R12 – Panel on Open Source NMT Toolkit Development
https://youtu.be/IjL3iLM6zJU
R13 – Dynamic Masking for Improved Stability in Online Spoken Language Translation
https://youtu.be/zwDZ4PBOBhs
R14 – On Target Segmentation for Direct Speech Translation
https://youtu.be/jVrspnywsuM
R15 – Domain Robustness in Neural Machine Translation
https://youtu.be/FtDZtzKMnkE
R16 – Low Resource NMT: A Study on the Effect of Rich Morphological Word Segmentation on Inuktitut
https://youtu.be/3UhrMrgXTBM
C1 – Operationalizing MT quality estimation
https://youtu.be/sBpb9CBYQUo
R2 – Investigation of Transformer based Latent Attention Models for Neural Machine Translation
https://youtu.be/OLQAhn43K14
C4 – COMET Deploying a New State of the art MT Evaluation Metric in Production
https://youtu.be/QZo35QTyOeQ
C5 – Scaling up automatic translation for software: reduction of PEMT volume with customer impact
https://youtu.be/QJekpmTB6-A
C6 – Auto MT Quality Prediction Solution and Best Practice
https://youtu.be/uGRnZVFYfdQ
C7 – A language comparison of human evaluation & quality estimation
https://youtu.be/LCaZN588SdI
C8 – Machine translation quality across demographic dialectal variation in Social Media
https://youtu.be/xaL0p_x2AVU
C9 – Making the business case for adopting MT
https://youtu.be/DAvIysaR7Qc
C11 – Enhance CX with Neural Machine Translation Technolog
https://youtu.be/lB9l5xXGm_0
C13 – Use MT to Simplify and Speed Up Your Alignment for TM Creation
https://youtu.be/S6ga5agfE20
C14 – Selection of MT Systems in Translation Workflows
https://youtu.be/wCvYMO6YQWY
C15 – Beyond MT Opening Doors for an NLP Pipeline
https://youtu.be/OEhNZ6-ZK3s
C16 – Building Multi Purpose MT Portfolio
https://youtu.be/c3Huf3g8MZo
C17 – Simultaneous Speech Translation in Google Translate
https://youtu.be/o55QArFJiJ8
C18 – Understanding Challenges to Enterprise Machine Translation Adoption
https://youtu.be/VCIJ-xqIYWs
C19 – Lexically Constrained Decoding for Sequence Generation
https://youtu.be/kHfh7WdP_T8
C20 – Building Salesforce Neural Machine Translation System
https://youtu.be/__C4odjXJUc
C21 – Interactive Adaptation of Neural MT on Commercial Datasets
https://youtu.be/-zd-wTTXjb0
C22 – Enabling New MT Post Editing Scenarios with Continuous Localization
https://youtu.be/Vr7Gro-Lw9o
AMTA Business Meeting
https://youtu.be/9VsGBR90fjo
T3 – NMT Domain Adaptation Techniques
https://youtu.be/v55Ax245AqU
W2 – iMpacT 2020 (Part 2)
https://youtu.be/bjkWwQYlZrc
W2 – Virtual Workshop on the Impact of Machine Translation iMpacT 2020
https://youtu.be/_TUoecpkCBY
Demo – Amazon
https://youtu.be/BH0O6Nuz4WE
Demo – XTM
https://youtu.be/NpbX7o2Ls5I
R5 – Machine Translation System Selection from Bandit Feedback
https://youtu.be/wS4LGBiitMg
W1 – 1st Workshop on Post Editing in Modern Day Translation PEMDT1
https://youtu.be/woXNepvWQaE
T1 – Quick Start Guide to Understanding & Working with Machine Translation
D1 Pangeanic Demo
D2 Amazon Web Services Demo
D3 Welocalize Demo
D7 STAR Group Demo
D8 Unbabel Demo
D9 Systran Demo
D10 Language Weaver Demo
D11 Intento Demo
D12 Apptek Demo
G1 Machine Translation as a Prototype for Advanced AI Deployment in Government
G3 You’ve translated it, now what
G4 Uplifting Singapore’s translation standards with the community through technology
G6 Thoughts on the History of Machine Translation in the United States
G7 Hand in 01101000 01100001 01101110 01100100 with the Machine A Roadmap to Quality
G8 Dragonfly Automated Sign Language Recognition ASLR and Machine Translation MT
G9 NVTC’s Transliteration Plug in What’s in a Name
G10 Robust Translation of French Live Speech Transcripts
G11 Speech to Text and Evaluation of Multiple Machine Translation Systems
K1 Welcome Steve Richardson & Keynote Address Marco Trombetti
K2 Keynote Address Alex Waibel
K3 Keynote Address Angela Fan
P1 Panel Recent Advances in Dynamically Adapted MT
P2 Panel Advances in Spoken Language MT
P3 Panel Large Multilingual Language Models and MT
R1 Building MT for Software Product Descriptions Using Domain specific Sub corpora Extraction
R2 Domain Specific Text Generation for Machine Translation
R3 Strategies for Adapting Multilingual Pre training for Domain Specific Machine Translation
R4 Prefix Embeddings for In context Machine Translation
R5 Fast Vocabulary Projection Method via Clustering for Multilingual Machine Translation on GPU
R6 Language Tokens A frustratingly Simple Approach Improves Zero Shot Performance of Multilingual
R7 Low Resource Chat Translation A Benchmark for Hindi English Language Pair
R8 How Robust is NMT to Language Imbalance in Multilingual Tokenizer Training
R9 How Effective is Byte Pair Encoding for Out Of Vocabulary Words in Neural Machine Translation
R10 On the Effectiveness of Quasi Character Level Models for Machine Translation
R11 Improving Translation of OOV Words using Bilingual Lexicon Induction
R12 Doubly Trained Adversarial Data Augmentation for Neural Machine Translation
R13 Limitations and Challenges of Unsupervised Cross lingual Pre training
R14 Few Shot Regularization to Tackle Catastrophic Forgetting in Multilingual Machine Translation
R15 Quantized Wasserstein Procrustes Alignment of Word Embedding Spaces
R16 Refining an Almost Clean Translation Memory Helps Machine Translation
R17 Practical Attacks on Machine Translation using Paraphrase
R18 Sign Language MT and the Sign Language Lexicon A Linguistically Informed Approach
R19 NMT Approach to Translate Text to Pictographs for Medical Speech Translation
R20 Embedding Enhanced GIZA++ Improving Word Alignment Using Embeddings
R21 Gender bias Evaluation in Luganda English Machine Translation
R22 Adapting Large Multilingual MT Models to Unseen Low Resource Languages
R23 Measuring the Effects of Human and Machine Translation on Website Engagement
R24 Consistent Human Evaluation of Machine Translation across Language Pairs
R25 Evaluating Machine Translation in Cross lingual E Commerce Search
Tutorial 1 Introduction to MT Presented by Jay Marciano
Tutorial 2 AutoML for Neural Machine Translation Presented by Kevin Duh
Tutorial 3 Executive Roundtable Chaired by Konstantin Dranch
Tutorial 4 Machine Translation User Guide 2022 Edition Presented by Konstantin Savenkov
Tutorial 6 Recent Advances in Translation Quality Evaluation presented by Lavie and Stewart
Tutorial 7 The Post editor Toolkit Presented by Luciana Ramos
UP1 PEMT human evaluation at 100x scale with risk driven sampling
UP2 Picking Out The Best MT Model On The Methodology Of Human Evaluation
UP10 A Multimodal Simultaneous Interpretation Prototype Who Said What
UP11 Data analytics meets machine translation solution
UP12 Quality Prediction
UP13 Comparison Between ATA Grading Framework Scores and Auto Scores
UP14 Lingua Addressing Scenarios for Both Real Time Interpretation and Automatic Dubbing
UP15 All You Need is Source! Source based Quality Estimation for NMT
UP16 Knowledge Distillation for Sustainable Neural Machine Translation
UP17 Innovations in Machine Voice for E learning and Training Content
UP18 Business Critical Errors A Framework for Adaptive Quality Feedback
UP19 A Snapshot into the Possibility of Video Game Machine Translation
UP20 Customization options for language pairs without English
AMTA Business Meeting
Conference Closing and Best Presentation Awards
Orlando, Florida, USA
12-16 September
AMTA 2022, the premiere Machine Translation conference in 2022, will provide engaging and productive networking opportunities for in-person attendees, as well as virtual access for remote participants from around the world.
The Machine Translation industry is in a fast growth and majority adoption phase, projected to exceed USD $500 million in 2022. Research is accelerating rapidly, and many bold and large-scale offerings have come to market recently with still more around the corner: machine interpretation for meetings, automatic dubbing of videos, speech-to-speech translation, machine translation for sign language and multimodal translation adopted by search engines.
This 15th biennial conference of the Association for Machine Translation in the Americas will be the most prominent MT event of the next year. Our previous AMTA 2020 and MT Summit 2021 events had the highest attendance in the almost 20-year history of the series, and we hope to break the record again in 2022.
We have waited more than two years to enjoy the spectacular Sheraton hotel venue in Orlando, Florida, located just outside the Disney theme parks. We anticipate that this venue will provide additional reasons for our diverse audience to come together, if they are able, and once again enjoy stimulating in-person networking opportunities.
AMTA conferences are unique in bringing together MT researchers, users, and providers of MT technology from academia, industry, and government.
For researchers, AMTA 2022 will provide unique opportunities to share their latest results with colleagues as well as understand real-world user requirements.
Participants from industry and government will benefit from updates on leading-edge R&D in Machine Translation and have a chance to present and discuss use cases.
As with our previous conferences, AMTA 2022 will provide parallel tracks of sessions addressing a variety of topics. There will be outstanding keynote talks and panels by recognized MT experts, fresh demonstrations of the latest offerings from MT providers, relevant tutorials for both beginners and more experience practitioners in MT, and in-depth workshops for specialist participants. Students interested in MT and computational linguistics will be able to connect with academic and industry mentors.
We hope to see many of you again online, but we especially hope to see many more of you in person in Orlando!
AMTA 2022 YouTube Recordings (Members Only) Main Conference Research Track Download(13MB) Users and Providers Track and Government Track Download(31MB) Workshops Workshops on Empirical Translation Process Research(WeTPR) Download(4MB) Workshop on Corpus Generation and Corpus Augmentation for Machine Translation(CoCo4MT) Download(2MB)
Orlando, Florida, USA 12-16 September AMTA 2022, the premiere Machine Translation conference in 2022, will provide engaging and productive networking opportunities for in-person attendees, as well as virtual access for remote participants from around the world. See More Information and Register Now Subscribe to conference updates and to receive an invitation The Machine Translation industry […]
