For us, quality is a blend of having a good initial text to work from, feeding it through our domain-adapted machine translation, and then intelligently distributing these outputs to a curated community of editors, who we support with tools and aids that allow them to review, post-edit and approve the content as fast as possible.
Unbabel’s Award-Winning Quality Estimation System
The Unbabel AI Research team (André Martins, Ramon Astudillo and Fábio Kepler) led the quality estimation experiments.
Dr. Helena Moniz runs the quality team at Unbabel on a daily basis.
Source: https://unbabel.com/blog/unbabel-translation-quality-systems/
- Absolute
- access
- Additional
- AI
- ai research
- aids
- Benchmark
- BEST
- build
- Campaign
- Campaigns
- care
- City
- closer
- Common
- Communications
- community
- competition
- content
- Creating
- Customers
- data
- delivery
- develop
- dublin
- editor
- EU
- FAST
- Feature
- First
- first time
- Fix
- full
- GitHub
- good
- great
- guidelines
- here
- How
- HTTPS
- huge
- industry
- information
- IT
- Job
- join
- Key
- knowledge
- language
- Led
- linguistics
- List
- Long
- machine translation
- measure
- model
- MT
- Natural Language
- Natural Language Processing
- network
- Neural
- neural network
- open
- open source
- Other
- Pay
- performance
- platform
- pool
- quality
- research
- Results
- review
- running
- Scale
- shared
- store
- Student
- studies
- success
- successful
- support
- system
- Systems
- Target
- Technology
- test
- Testing
- third parties
- time
- track
- Training
- Transactions
- Translation
- Translation Quality
- Translations
- Unbabel
- university
- us
- vision
- Voice
- weekly
- WHO
- within
- words
- Work
- world
- worldwide
- year
More from Unbabel
The App Guy Podcast: Vasco Pedro shares his advice on how to successfully grow your startup
Source Node: 826711
Time Stamp: Mar 8, 2016
Tired of bad machine translation quality? Get to know Unbabel’s human translation API
Source Node: 823932
Time Stamp: Apr 10, 2018
How telecoms giant Italtel stopped worrying about translation and localized its content with Unbabel
Source Node: 825926
Time Stamp: Feb 10, 2017
#Salvadorable: Understand Portugal’s first-ever Eurovision winner in 17 languages
Source Node: 825670
Time Stamp: May 15, 2017
By the time you start thinking about churn, it’s too late
Source Node: 799058
Time Stamp: May 9, 2019
How languages die: 3 dead languages and what they can teach us
Source Node: 804391
Time Stamp: Dec 6, 2018
Japanese, Finnish or Chinese? The 10 Hardest Languages for English Speakers to Learn
Source Node: 828985
Time Stamp: Mar 6, 2015