Author: Weixin Mo

Video Localization Project

Video Localization Project

Subtitling and Localizing a video

Introduction:

My Desktop publishing (DTP) final project in the 2018 Spring semester is subtitling and localizing a Chinese commercial advertisement video by using Adobe After Effects. It is a collection of several advertisements produced by the same web game manufacturer. These ads are some of the best-known ads in China, especially by the young. They are famous mostly for their hilarious conversations, awkward pronunciation of Chinese, and garish graphics. They have been accumulating popularity at an amazing speed in China and produced some famous internet memes as well. The source video is in Chinese, but the original subtitle is at very low quality, incomplete, and the font styles are inconsistent, as shown in the screenshots below.  Continue reading “Video Localization Project”

CAT Final Project Portfolio

Introduction:

This Statistical Machine Translation(SMT) Project is to train the Microsoft Translation Hub, which is the SMT provided by Microsoft, to assist human translation of the subtitles of Marvel superhero movies. By the end of our training, we have completed 22 rounds of training in total, with 129,869 segments added into the machine. The updated proposal aims to present the analysis of the pilot project, the estimates of the complete project, and recommendations for how to adopt machine translation in your company’s translation workflow. Continue reading “CAT Final Project Portfolio”

Translation Management System Project

Our project is to translate a commercial document for Clarins, which is a worldwide known luxury cosmetic manufacturer. We use World Server as our Translation Management System (TMS), and we also run a round of pseudo translation in World Server as a test run. The source file is a brochure, We localize the original file from English to Chinese, then do the Desktop publishing to makes a perfect Chinese version of the brochure. The deliverables include:

  1. Source file
  2. Target file
  3. Pseudo Translation document
  4. Term base
  5. Translation memory
  • Our proposal can be downloaded here
  • The deliverable can be downloaded here
  • Our final presentation can be downloaded here
  • A blog post with detailed introduction of selection of Translation Management System(TMS) is here
  • I also make an infographic regarding the TMS comparison, it can be downloaded here

 

TMS Selection and Recommendation

Why do we need a TMS?

A Translation Management System (TMS) can do the basic and repetitious works that may be tedious for human workers. Generally, TMS has its own automation system to facilitate a project. Project Managers can apply the workflows that are provided by TMS, so that they do not need to spend time on making new workflows, and hence boosts the working efficiency. TMS can also reduce human errors since many tasks, such as sending notification, can be done by its automation system.

TMS Recommendation:

I would recommend Lingotek as a wonderful model of Translation Management System. Firstly, Lingotek has a very user-friendly interface. The interface is clean and straightforward, with the tool bars located on the left side of the screen. The system is high automated since most of the resources can be synchronized with the projects. Lingotek has a very useful search panel as well; it will not take so much time for project managers to search for their resources.

Secondly, Lingotek gives the users much freedom to manage projects. It allows the users to customize the workflows themselves. Users can customize, add, or delete phases in a project workflow based on their needs. In addition to it, they can also make adjustments for each of the phase, so the entire process is flexible and highly automated. Lingotek also allows users to do translation on its embedded workbench. Translators can have real time editions on the workbench, which is very convenient.

In addition to what I mentioned above, users can also create and manage linguistic tools on Lingotek. For example, users can create Translation memories and Term base in the browser, can these data set can also be downloaded and further imported to CAT tools, such as SDL Trados.

Finally, Lingotek is cloud based. It is an online system, thus the users do not have to install a new application on their desktops. The entire system is easy to access, while people may have to pay for buying license if they want to use some other TMSs.

Experience with XTRF

The needs for translation and localization have never been so strong. As a fast-increasing number of enterprises and companies are seeking for expanding their sales scope oversea, translation and localization begin to play vital roles. However, translation has no longer been a simple process——one people gives you the source file and then you give him back the translated file. This is not in a translation class, in which all you need to do is translating the document that your instructor gave you and then turned in. Instead, translator is a part of the vendor team, and plays a role in a complicated workflow. Clients, sales person, project manager, language provider and many other roles work together in a project, thus translation management systems are necessary to automate the workflow and reduce the manual processes which may be tedious or time consuming.

XTRF is a TMS that manages the flow of globalization and localization processes. The fundamental tasks throughout the translation are automated. It also has different portals for people in different positions, and they are simultaneously connected so that people, such as PM and vendors do not have to meet together to get things done. For me, XTRF is a convenient tool to have a clear overview of the whole workflow, and I will talk about my experience of using it from several aspects.

Interface:

The interface is very user friendly, as shown on the picture.

It is user friendly because roles and task are clearly categorized, so that it does not take much effort for the project managers to tell where to click. 

Man Management:

The project manager can view, add, and edit the clients’ information, assign the contact persons by clicking the button “clients”. Clients can be added here or clicking the “Add” tab on the top of the interface.

The data of the vendors are pretty detail. Other than the personal information, such as contact info and language combinations, CAT tools, software and hardware using are also shown, which is very convenient for the project managers, helping them to decide which vendor can be contacted to complete a certain type of project.

In addition to those above, XTRF also has a wonderful email management. Emails can be sent automatically, thus the project managers do not need to manually send notifications to all of the people involved. This feature will be even more important if a Project manager is in charge of a complicated project, for example, to translate the source file into 30 languages, and more than one vendors may be needed.

In short, XTRF knows the workflow well, so it manages to send notification whenever an important further action is needed; for example, asking for approval of quotations, notifying the translators to start, telling the proofreaders to start reviewing, and so forth.

(The list of vendors)

Workflow:

XTRF is a highly automated system which frees the PMs from doing much tedious works. For example, after the creation of a project, a quote can be automatically compiled and sent, with almost all the information that the clients may use to approve a quote. In addition to it, PMs can supervise and see the progress of the current project, without having to worry about losing control.

Another fantastic feature is that XTRF allows the project managers to customize the workflow base on their own preference; and it can also be adjusted during the project, hence users do not have to start from the beginning.

Compatibility:

XTRF supports a large number of file formats. Commonly use file formats, such as TMX and XLIFF are supported. However, XTRF is not a CAT tool, so it is not a platform for translation.

Conclusion: 

As a conclusion, XTRF is one of the most fascinating Translation Management Systems. If you are a project manager, XTRF can not only enhance your working efficiency with its highly automated system, but also provide the users high quality management of the localization process. If you are a vendor or a client, this TMS can also prevent you from wasting much time on looking for contact persons, unnecessary communication, and thus boosts your time efficiency.

Website Localization Project

Our Project is to localize a real website by using transifex, a proxy translation platform. To avoid any copy rights issues, we decided to choose the website of a non-profit organization, and the final translated website will not be published. In this project, we use Plant with purpose as our source website. Transifex requires the users to choose the type of the projects, the files-based project is for localizers who have the website pages on files, and the Live Project option is for those who want to localize the website in real time through JavaScript. Transifex allows a team to administrate a project, which is convenient for a team to collaborate.

After adding the website URL to transifex, the system automatically generated a report, however, the word count report was incomplete because we had not yet make scans. Transifex counts detected strings only, and the word count report will be updated every time when the system detects new words or strings; moreover, the words showed on the report will be counted as using the “quota”, thus the users do not have to worry about wasting money on translating the words that they do not want to translate. This is different from Easyling (Easyling will count every word found in the first scan, so it is very hard to not exceed the word limit).

Translation memories can be imported and exported. This function will be particularly important if the translators need to convert the source language to tens of target languages. Each language will possess a unique report.

The most fascinating function in transifex is the live translation, allowing the user to apply a live preview of the translated website. When going into the live translation mode, the system will automatically run a scan through the current page, determining the strings to be translated. After the scan, users can approve the detected strings or exclude some of the strings, while approved strings will be ready for translation. All the translation jobs can be done in the editor. In this case, we used Google translation to machine translate all the approved strings. Marking the translated strings as “Reviewed” is all a user needs to do to finish the translation tasks. The live translation editor can be regarded as a small CAT tools, since we are doing the same thing as we will do in SDL Trados. No coding is needed, thus, even a translator without any relevant knowledge in programming can finish this on his own.

As mentioned before, transifex only run scans on current page, which means that following every link in the whole website and translating it separately is necessary. Links cannot be directly followed like what we do in a real website. Users need to click the button “follow the link” because a single click on the link is to edit the text, not to follow it. All of the translated pages can be chosen and reviewed by following the commands in the right tools bar.

The final job needed to be done is to publish the translated website. Transifex provides the project administrators with the JavaScript codes to show the translation. The project admins need to copy the codes and paste them onto every page that needed to be localized. Then, transifex also provide a staging server, which is staging.transifex.com. By using the staging server, localizers are able to test the project without publishing to the public.

Overall, transifex is a powerful and user-friendly proxy translation tools. It is easy to use, and the whole system is highly automated. It is definitely a very good choice for localization companies. However, its drawbacks are also apparent. The first one is the price. Transifex is much more expensive than its rivals, as its cheapest subscription is 179 dollar/month, with only 50K hosted words available. If a company does not have lots of website localization business, transifex is likely to be a bad choice. Second, although more than one user can localize the website at the same time, transifex sometimes crashes if two workers doing the jobs at the same time, especially if they are working on the translation for different pages, the auto detection may produce some bugs.

Crowdsourcing Translation —— Quality control

Some common worries about crowdsourcing may deter people’s interests. A typical one is that how do companies maintain their quality-control (QA) for crowdsourcing translation.

  1. The first practice is to hire LSPs to complete the final reviews of translations, like what Facebook did for their website localization into hundreds of languages. It is a fairly direct way to keep high-quality translation, and do not take much effort for the companies of reviewing, as they hand over the tasks to language providers. Another merit is, professional language providers can help the companies to catch up the deadlines.
  2. An alternative way to maintain QA is to employ a voting system; It is expected that translation with the highest votes will be more likely the better translation. Many translation management systems allow the project managers to choose voting as the way to accept translation. Typically, project managers can decide for how many votes that a translation will be automatically accepted by the system.
  3. For companies which want to localize their products to hundreds of languages, a difficult challenge will be that it is sometimes hard to find enough translators, not to speak of professional reviewers. In such cases, project managers can apply the round robin test——letting each of the competitors to grade all other people, in order to decide the potential best translator.

 

Crowdsourcing Translation

Crowdsourcing translation is an effective way to tap into the expertise of users. The advantages of crowdsourcing translation are very visible: It can be an expeditious way to work on a translation project, since the number of workers can be indefinitely large, and they are always available. It can also build a bridge between the company and users, enhancing the brand loyalty. However, despite that crowdsourcing seems constructive and beneficial, a difficult question may arise —— How can I find so many volunteers? Even if I find them, how can I keep them motivated to do the translation for us? Crowdsourcing usually targets large numbers of people. It is true that if project managers fail to have enough volunteers, the word “crowd” will not be suitable for crowdsourcing translation. Thus, in this blog post, I will focus on how to maintain quantity.

  1. One of the most effective ways to make people happy and willing to work is to give them presents and rewards. For example, giving them memorial T-shirts which can never be bought elsewhere is practical. Volunteers who want to be recognized as one of the contributors, or someone who simply just want to have a nice limited-edition T-shirt, are likely to be motivated. However, understanding what rewarding system is suitable for the case is important. Volunteers may have different preferences or expectations for their rewards. According to Jeff Beatty, who is the Head of Localization at Mozilla, he said volunteers in Western Europe tend to seek for self-satisfaction, the sense of being needed, and the contribution behaviors themselves are quite enough for them to keep motivation; while American volunteers will prefer T-shirts, caps, things that can be used to let others know they participated. Asian volunteers like food; free snacks or pizza will likely be ideal for motivating them. Therefore, project managers should consider using various kinds of rewards for people with different culture background.
  2. Keeping in touch with your volunteers also plays a vital role in maintaining motivation. People do not like to be left alone with no further feedback. Companies and organizations need to constantly give them feedback, such as telling them that their translation have helped the project to go smoothly, asking them to keep an eye on what will come in the near future. Emails and calls are necessary.
  3. Machine translation can also be used to motivate volunteers. Many people have the experience that starting is the most difficult part for doing things, but after the starts, things go much more smoothly. Machine translation is like priming the pump; it helps to start projects and volunteers will not be intimidated to start and do not have to worry about where to begin.
  4. Encouraging people joining the programs usually requires good marketing. The four Ps of marketing is a widely used model. It consists of Product, Price, Placement, and Promotion. For the products, project managers need to tell the potential volunteers that what they can get. It can be wonderful experience, good feelings, or helping them to build constructive social connections. For the price in this case, the opportunity costs may not be too high for the volunteers, since it will deter people’s interest. The organizers need to cooperate so that volunteers will be easy to join. Some typical practices are: having a flexible schedule, providing guidance to the tools; and if organizers are holding a meeting event, then they should think about offering children care. All of these efforts are aiming to reduce the costs for the volunteers. Placement should not be neglected as well. Project managers and organizers need to find a place that is easy to commute. As for promotion, advertisements can be posted on social media, like Facebook groups, Instagram, LinkedIn, etc. Local newspaper is also a good choice if targeting elder volunteers.

Quantity is essential: time spans, quality, and consistency are all heavily affected by quantity. Lacking enough users and volunteers will expectedly defer the project’s accomplishment. The practices elaborated above are some typical ways, more practices are also needed based on the actual situations. People may argue that even if we have many volunteers, how can we ensure that the translations are high-quality enough to meet the requirements? Here is my personal opinions regarding quality control for Crowdsourcing Translation.