Companies are reconsidering their traditional localization and translation strategies. The classic consideration required making trade-offs between volume, quality, and cost, generally forcing decision-makers to choose two of those three. This led to a fragmentation of translation options: human or computer, from high-quality to low.
- If you want high volume with high quality, you have to accept high costs of a legion of high-value human translators.
- Those focused on high quality, but with limited budgets, also chose human translation, yet significantly minimized their translation volume.
- Those requiring high volumes focus on Machine Translation (MT), and accept a trade-off in quality. Even the best MT solutions without human editorial review are underwhelming in terms of output quality.
- Another choice to economize requires risking your enterprise upon low-quality/low-cost human translation providers, or the uncertain proclivities of crowdsourcing. Neither of these paths are truly as low-cost or free as they seem, as they require significant internal editorial review and management due to the resultant high variance of quality.
- For many facing no optimal solutions, it is simpler to just hold off and skip translation altogether.
A New Option: PEMT
Now there is another option: Post-Editing Machine Translation (PEMT). PEMT is the empowerment of high-quality human reviewers and editors with advanced Machine Translation (MT) technology — including the most recent autoadaptive Machine Learning (ML) systems. PEMT may involve multiple steps, including pre- and post-processing of content.
The net result is that the human reviewers produce far higher volumes of content, at a fraction of the cost, with nearly the same level of quality. PEMT can save customers up to one half (or more) of the price of traditional high-quality human translation services. Or, if you’ve been used to machine translation alone and have been unhappy with the results, watch your translation quality rise dramatically with a marginal increase in price.
Also keep this in mind: not every translation service providers’ offering of PEMT is the same. e2f is headquartered in the heart of Silicon Valley. We work with the latest technology providers and research developments to stay abreast of the ever-evolving fields of machine translation, machine learning, and translation automation practices.
Relative Price, Quality and Volume
|Translation Type||Relative Price||Relative Quality||Relative Volume|
|Post-Editing Machine Translation (PEMT)||$$ to $$$||☆☆☆☆||????|
|High-Quality Human Translation||$$$$$||☆☆☆☆☆||??|
|Machine Translation (MT)||$0 to $$$||☆ to ☆☆☆||?????|
|Low-Quality Human Translation||$ to $$||☆ to ☆☆☆||????|
|Crowdsourcing||“$0”||☆ to ☆☆☆||? to ???|
Post-Editing Machine Translation (PEMT) offers nearly the same quality as the best human translation, with a significant reduction in price, while maintaining the high volumes of MT. PEMT can be accomplished at half the cost (or less) of comparable human translation.
High-quality human translation is the best quality at the highest price, but cannot easily scale to the highest volume jobs.
Machine Translation (MT) options range in price and relative quality. While there are free consumer MT tools available, they often limit the volume they translate at no charge, and result in variable levels of quality. High-volume MT actually has a cost. By itself, MT cannot currently meet many standards of quality.
Low-cost/low-quality human translation results vary widely; often little better than MT. It “scales” only because there is a large pool of less-talented translators in the industry.
Crowdsourcing, while ostensibly “free” requires significant management and review processes. Volumes that can be managed by crowdsourcing, and the resultant quality, are both variable.
If you would like to incorporate PEMT into your localization strategy, contact us at firstname.lastname@example.org to discuss your questions, needs and vision.