The Art of Hyping Machine Translation
What’s behind the arcane research paper that Baidu helped spin into a machine translation breakthrough celebrated across the media?
arXiv
Google, Microsoft, Baidu Step Up NMT Research in Record August as BLEU Takes Heat
August 2018 became the busiest month ever for neural machine translation research as more papers call for improved evaluation methods and big tech companies step up the number of submissions.
TWB, Big Tech, Academia Take on Covid-19 Language Challenge
Collaboration between Amazon, Facebook, Google, Microsoft, etc., and Translators Without Borders aims to fight Covid-19 infodemic through machine translation research.
Reader Polls: Unpacking ‘Human Parity’, Public Sector Demand, Startups, Growth Regions
Slator readers give their opinion on public sector demand for language services and technology, the impact of startups on the industry in the next three years, the highest potential growth regions, and how neural machine translation should be evaluated.
In Human vs. Machine Translation, Compare Documents, Not Sentences
Microsoft’s “human parity” claims drew skepticism from a team of neural machine translation researchers as they found that much of the research and industry community is looking at the problem from the wrong angle.
Slator Pro Guide: Finding Growth in Adjacent Services
Pro Guide for LSPs to identify growth opportunities in adjacent services. Profiles 15 adjacent services, analyzing barriers to entry, risks, and synergies.
Google Ups Ante in Speech-to-Speech Translation With ‘Robust’ Translatotron 2
Google researchers say Translatotron 2 outperforms original in terms of speech-to-speech translation quality and naturalness. Plus the new version prevents users from generating speech for deepfakes.
TikTok Parent Company ByteDance Open Sources Neural Speech Translation Toolkit
Chinese tech company ByteDance (TikTok) hopes NeurST, which bypasses the transcription step integral to traditional, cascaded speech translation, will standardize benchmarks in the field.
What Makes a Good Post-Editor: Research Examines Activity Patterns of over 300 Linguists
Characterizing post-editors based on their keystrokes, mouse activity, and other editing actions can help predict speed and, eventually, quality, says a new paper by Unbabel researchers.
Terminology Assistance Coming to a Simultaneous Interpreter Near You
Real-time terminology assistance for simultaneous interpreters may be well on its way — but will it be truly useful in typical high-stakes environments?
Behold, the Translatotron! Google’s Latest Move in Speech Translation
Google researchers approach speech-to-speech translation a little differently in a proof-of-concept that completely takes away reference text, but underperforms compared to existing models.
Corporates Going All in on Neural Machine Translation Research
Long the domain of academia, fundamental research in machine translation has become a battleground for big tech and niche providers.
The Race for Open Source Neural Machine Translation
Systran’s Global CTO Jean Senellart discussed the direction the open source race for neural machine translation is taking and what it means, what’s its bottom-line, and what to expect in the future.
Reader Polls: Is NMT Impacting Pricing? Subcontracting, GDPR, Media Localization
Slator readers tell us what they think about buyers prohibiting subcontracting to other providers, the media and game localization boom, Europe’s General Data Protection Regulation rollout, and a resurgence in neural machine translation research and its ongoing implications.
Google, Facebook, Amazon: Neural Machine Translation Just Had Its Busiest Month Ever
Research into neural machine translation in 2018 is exploding, with April 2018 being the busiest month ever for NMT, as Big Tech companies from Google to Facebook to Amazon go full throttle on multiple areas of research.
Facebook Wants to Understand the Uncertainty in Neural Machine Translation
Facebook’s AI Research team delves into the neural machine translation black box and figures out a couple of ways to mitigate performance degradation due to intrinsic and extrinsic “uncertainties.”
Here’s What Machine Translation Researchers Are Geeking Out On
Making do with less training data, infusing models with more document-level context, and increasing efficiency and reducing power consumption have been high on the agenda of neural machine translation researchers over recent months.
Neural Machine Translation Research Accelerates Dramatically
As interest in neural machine translation reaches new heights — with even Mark Zuckerberg posting about it on his personal Facebook page — academia continues to churn out NMT papers at a record clip; 2017 output already on par with whole of 2016.
In Research, Neural Steamrolls Statistical Machine Translation
Research into statistical machine translation is fast falling out of favor among the research community as the number of papers published on neural machine translation skyrockets in 2016.
Your information will never be shared with third parties. No Spam.