In late 2011 the TAUS Dynamic Quality Framework (DQF) working group published its first report on translation quality evaluation. This report kindled a strong interest in translation quality management. In early 2015 Willem Stoeller created a new training course on translation quality management and DQF in close cooperation with TAUS. This training presents a comprehensive overview of a translation quality methodology combined with a review of the DQF tools for content profiling, quality evaluation, machine translation engine comparison and post-editing productivity. The course also includes a number of real-world case studies. All participants will receive free access to the TAUS DQF tools for ten weeks. This will allow them to complete the exercises detailed in the course materials. Upon completion of the exercises, participants will receive a certificate of completion from TAUS and the Localization Institute.
- To better manage the issues related to managing translation quality
- To appreciate how different quality management standards can contribute to managing translation quality
- To be able to place the TAUS Dynamic Quality Framework within the context of quality management models
- To be able to use the DQF resources to guide your evaluation decisions
The target audience for this online course includes anyone interested in translation quality. In particular, translation service providers or buyer project managers, quality managers and lead translators, but also any other translation professional interested in translation quality using machine or human translation. This course is not intended for those who are new to translation or localization. In fact, it is recommended that you have at least three to five years of experience in the translation industry or an academic background in translation/localization.
- Overview. We’ll start with a review of what translation quality is, then will cover the project management standard ISO 21500 2012 and the translation quality standard ASTM F2575 – 14.
- Dynamic Quality Framework. This provides a look at the initial findings of the DQF group as covered in their 2011 report.
- Quality management methodology. Here we will review translation quality as a risk management problem. Translation requirements, quality evaluation methods, sampling, translation performance reporting and vendor balanced scorecards will be covered. This topic also includes a quality management case study that describes the design and implementation of a translation quality management process at VMware.
- Content profiling. This describes one of the components in DQF; the content profiler. The importance of content profiling will be reviewed followed by a demonstration of the use of this content profiler.
- Quality evaluation. Three of the most frequently used methods for measuring translation quality in a translation production environment — error typology, adequacy and fluency — will be covered. These methods can also be used when comparing the translation quality of different providers or machine translation engines.
- MT engine comparison. Several evaluation methods used by human evaluators to compare different MT engines will be covered. The same methods are also used to determine the impact of additional training of a statistical MT engine.
- Post-edit productivity. Here we will look at the time and cost savings introduced when using machine translation combined with post-editing.