Tag Archives: translation

Why shouldn’t journals publish translations of articles alongside the English version?

Update (14 April 2022): If you’re interested in the notion that something is lost when we default to English as the language of scientific communication, you may be interested in this reflective podcast by Shaun Cleaver that was prepared as part of the 2020 In beta unconference.

A few days ago I received a submission to OpenPhysio from someone who was clearly a non-English first language speaker. After a few rounds of email to make sure I understood the general structure and claims of the article, I decided that we’d go ahead and work together to tidy it up a bit, before sending it out for peer review. I know that reviewers can sometimes take on an editorial role as part of the process and wanted to make sure that the central ideas were clear.

However, it occurred to me that this may also be an opportunity to offer the author the option of preparing a translation of the article in their home language, to be published alongside the ‘original’ i.e. the English version. Authors go to a lot of effort to translate their work into English, which has this weird side-effect of closing it off to a population of non-English speakers, who may nonetheless have benefitted from reading it. I can only see upsides to this practice and almost no disadvantages, other than it adding a bit more work to the publishing process. And of course, authors would have to agree to take on the translation themselves (I’m talking from the context of a fee-free journal, like OpenPhysio, that wouldn’t be able to pay for this service).

There are no technical limitations that would prevent this. Making a second version of the article available is as simple as providing a link to the file. To start with, we could even say that the translation will be available as a ‘stripped back’ version, with no formatting and design i.e. it could simply be a PDF with the the original citation that points back to the canonical (English) version. Of course, the author can do this anyway but I think that making it available alongside the original would add some ‘credibility’ to the translation. This first iteration would just be a proof of concept. You can imagine that, over time, you could have it available in HTML (to help with discoverability), and also assign a DOI to the translated version to differentiate it from the canonical version. And you’d need to have a translator verify that the articles are the same.

I can’t think of any reasons for why we shouldn’t do this.

The researchers started with 140,000 hours of YouTube videos of people talking in diverse situations. Then, they designed a program that created clips a few seconds long with the mouth movement for each phoneme, or word sound, annotated. The program filtered out non-English speech, nonspeaking faces, low-quality video, and video that wasn’t shot straight ahead. Then, they cropped the videos around the mouth. That yielded nearly 4000 hours of footage, including more than 127,000 English words.

After training, the researchers tested their system on 37 minutes of video it had not seen before. The AI misidentified only 41% of the words… That might not sound like a lot, but the best previous computer method, which focuses on individual letters rather than phonemes, had a word error rate of 77%. In the same study, professional lip readers erred at a rate of 93% (though in real life they have context and body language to go on, which helps).

Source: Lip-reading artificial intelligence could help the deaf—or spies | Science | AAAS

There’s not much else to say here, other than to highlight one of the potential applications in healthcare. For example, patients who are hard of hearing could have a universal translator with them at all times. In a country like South Africa where we have a Constitution that mandates the provision of healthcare in a language of the patient’s choosing, but where we have 12 official languages and a huge shortage of translators, you can see how this might be useful.

Google Translate: when small things make a big difference

I’ve never really had the need to use Google Translate. Most of the content I come across is in English, and if it isn’t I’ve never read it and so never realised what I was missing out on. Earlier today I came across Ilona Buchem via a post from Stephen Downes, pointing out a presentation on PLE’s that she’d shared on Slideshare. The presentation led me to her blog, which was in German.

Chromium helpfully popped up a tab asking if the text should be translated from German into English, which I agreed to. After reading the first few posts, I decided that this was someone to include in my network. After clicking the RSS icon in the browser, it automatically imported the feed to my GReader account, again asking if this feed should always be translated into English.

Every so often I’m blown away by an elegant and intuitive solution to what was previously (for me, in this instance), an almost insurmountable obstacle.