|  ext and graphics can be challenging enough to localize. But with multimedia, you may be looking at a more complex puzzle including sound, animation and video. More than 60 people attended the SimulTrans Localization Seminar, "Localizing Multimedia" to learn just how complex the process is &emdash; but also how rewarding.
Speakers were Glyn O'Leary, managing director of SimulTrans' Dublin subsidiary, and Pat Wylie, localization coordinator for Humongous Entertainment in Bothell, Washington. O'Leary laid out all the pieces of the puzzle, how to localize each one and how to put things back together again. Wylie shared his experience at Humongous where he heads the efforts to localize video games, including such popular titles as "Putt-Putt," "Freddi Fish" and "Pajama Sam." Grasping the Big Picture "The complexity of multimedia development from a localization point of view is far greater than regular apps, in terms of ease of localization, cost, etc.," O'Leary warned up front. "It's one big puzzle that has to be taken apart and, quite often, put back together again almost from scratch. It's complicated and, technically, very challenging. It can also be very costly." Based upon the size of the applet, it can cost up to five times as much to localize a multimedia product as opposed to a standard application. Multimedia applications are also subject to "maverick development," according to O'Leary. Even in the same company two developers may do things quite differently. "This means that every time you take the puzzle apart, it's a brand new task." O'Leary has also encountered these obstacles: He tries to contact the developer of an application, only to find that he or she has moved on. The actual engine used to develop the product is "homemade" or indigenous to this particular development company (and there are thousands of multimedia development companies). The engine is loosely based on some commercially available engine and the company claims to have a Director application (which borders on being a breach of the Trade Descriptions Act). Because of competitive restrictions, the developer may also simply not want to offer specific and proprietary information. Engines will always pose problems because they are numerous and specific. "If you're doing 100 multimedia products, you'll probably come across 100 variations of engines," O'Leary noted. "And text will be embedded everywhere." Applications are routinely developed on one platform (often Macintosh) and then ported to another (often Windows). While it looks good in the development environment, it falls short in the running environment, which is where localization takes place. The application may even look good running on conventional platforms but fall apart on French Windows. No Way to Leverage With a routine Windows application, you can take advantage of some efficiencies as new versions come out. "You already understand how the app builds, how it installs, etc.," O'Leary said. "You can leverage the text." This, however, rarely happens in multimedia. "What you get back is the very latest technology that was developed with a different version of the engine, or a brand-new engine, or a brand-new developer." "We're dealing in a very competitive market. Survival means having the most awesome products," O'Leary noted. "The cool, new features, however, don't often play well in France," O'Leary continued. "You go from localizing an English product for France (or Germany or Italy or Spain), to a situation where you're nearly developing an indigenous, per-country product from scratch." Playing the Sound Card Localizing sound and video can drive costs up more, but are essential in the target market. "Italians would rather see Italians speaking, not a dubbed version," O'Leary said. "Furthermore, there are no neutral French or German accents; people may or may not like the accent you choose and complain, 'It's too industrial &emdash; too artistic &emdash; too whatever.' Complications relative to these two multimedia elements are enormous once you cross the Atlantic." According to O'Leary, the cost to have people put their voices to products runs up to $3,500 a day. Localizers can also run into a certain type of legacy. Large, blue chip publishers like Disney are synonymous with high quality. When Disney releases an application that follows a major film release, nothing short of the original film voices is acceptable for the characters in the multimedia application. Once a voice talent knows he or she is the only viable supplier, the price goes up. "The more specific the voice is in the US product, the more specific it will need to be in your target language," O'Leary said. "And the more difficult and costly it will be to secure. Multiply this by the number of target languages and you're talking real problems." As with text, sound can expand in another language by as much as 50%. "Taking sound into another language, say German, can completely throw off synchronization," O'Leary said. "What you end up doing is compressing the sound and dealing with the pitch." Synchronization is the single biggest issue in O'Leary's eyes. "People paying for the localization project do not want to have to re-shoot expensive video. Instead, they ask you to use the same characters, just put in a different sound. "Unfortunately, if we leave the video as is and put German in its place, the characters finish speaking five seconds after the video." If a German watches a video portion of a multimedia application that's out of sync, the drop in quality "will be very annoying and, in most cases, unacceptable." O'Leary suggests using animation when possible. "You don't have to sync in with anybody's lips," he explained. "You simply speak, take your time and lay it over." Graphics: Handle With Care O'Leary suggests that all graphics be stored in the industry standard, Photoshop EPS. The format allows for a file to contain various layers that can be translated, one by one. Also, don't use background gradation, rather, standardize and retain palettes. "Otherwise, expect a significant increase in time and money." As for fancy text and fonts, the rule is: The better it looks, the more costly it will be to replicate. "A localization engineer may need to take a graphic, cut out the English text, rebuild it with translated text, recreate the same fonts and fill in the various colors. It might take hours and, as a result, incur significant costs per screen. And you could be dealing with hundreds of those screens." Humongous Up to the Challenge Humongous Entertainment, founded in 1992, produces multimedia games for children ages 3 to 10. The company's multimedia products consist of animation, "talkies," music and high-quality, hand-drawn graphics. It also produces hybridized games into Windows 3.1 and 95, and for Macintosh. The company has localized some 10 games per year on average and is expected to complete 40 during 1999. Pat Wylie, Humongous' localization coordinator, works with the developers to make sure that the international market is considered during development. "I encourage developers to look at US products with a localization mindset," he explained. "The development cycle is never long enough and, once the crunch mode starts, thoughts of localization can go out the door." Because the programming language Humongous uses is proprietary, the code can't leave the building. "This makes it difficult for a company such as SimulTrans to localize our products," Wylie said. Items that do go out include text for the talkies, art and packaging. "This calls for extensive preparation since our localization partners receive limited items," Wylie said. "If our information is poorly thought out, it requires more time to localize." Building a Better Kit Humongous' localization kits "include everything other than the proprietary language, such as maps which reveal which parts of the game need to be isolated," Wylie said. "We have character descriptions and scripts which allow the actors to get a feeling for what's going on with the characters, how they can suit their acting style and turn it into German or French, for example." Wylie's department also drops into each kit art and text, Help files, read-me files, resource scripts for dialogue boxes and packaging. The kits typically take one to two months to prepare and Wylie makes certain they start to take shape during the US production cycle. His department must hand off to its localization partner some sort of reference for what the translators will be working with. "To do so, we make a composite frame, put the graphics together, then forward it to our localization partners who translate the materials and return them. We then strip off the back graphic." Unfortunately, the process is problematic for Wylie and his team for a number of reasons. "It can take weeks when you're dealing with 500 frames of art," he said. "We're going to scrap this process and do the art in-house. We'll simply give our partners text documents to translate, then place the text in the animation files." He figures that under the new method, two people working full-time can complete 40 games in one year. "Because our games are rather big and have somewhere around 8,000 talkies and 2,000 animations, they make for cumbersome localization kits," he said. Someone on Wylie's team systematically goes through and looks at each frame to make sure there isn't any text in it. If there is, it needs to be saved. "It's something to consider if you're a developer: document everything." The Essential Script Wylie views scripts as the most important item in his localization kit because they allow the actors to see the references and get a feeling for what the characters meant in English. "They can then convey the same feeling in their respective countries," he explained. His games feature two scripts containing the same lines or talkies, one a Word document, the other an Excel spreadsheet. In the initial Word script, items are organized by room, line number and character, which allows for flow of conversation in a room. "We also add contextual references for lines that are puns, jokes or have cultural meanings," he said. Because Humongous' games are filled with puns and jokes, contextual references can be found above nearly every line of a talkie. In the Excel spreadsheet script, items can be sorted in a variety of ways, i.e. by character, talkie number or room. This allows people at the recording studio to pull out identical talkies that occur in different rooms. "We also re-use numerous talkies or portions of them because of high recording costs," he added. "This spreadsheet allows people to cross-reference and figure out which talkie happens in multiple places." Wylie is quick to point out that such procedures take a great deal of time and cost a great deal of money &emdash; but it's worth it. "Multimedia games cost a lot of money to produce," he said, "but they also sell well. We have about 8,000 talkies maximum in a game, which can cost up to $50,000 to record." Lip Syncing and Its Costs Lip syncing is expensive. Wylie's team tried to lip sync one game in its entirety and had to hire an additional staff of 10. Because of the high cost, Humongous tries to avoid lip syncing except for extreme close-ups and facial shots. Instead, Humongous relies upon an animation suite that features tokens and markers. This allows the character to "start flapping its mouth or have certain mouth openings," Wylie explained. His company also uses a random mouth-flapping generator that allows a talkie to go longer if necessary, as in the case of German. "Because of these options, we don't have to do any expanding or any shrinking," he said. "We don't have to ruin the sound of the talkies. If you start stretching and compressing 8-bit, 11-kHz sound, it will sound horrible."
 Putt-Putt becomes Pounce-Pounce in French
Final Integration and Testing Wylie's team integrates the art, talkies, resource script and dialogue boxes. The sound effects and music are first because, invariably, out of 8,000 talkies, one goes unrecorded. "Our compiler detects this and we're able to instantly tell our partners who can then book studio time to record missing lines." His company gives him two months to integrate and test. If Wylie is short of resources or assets for the game, "it's going to take our localization partners a month to secure recording time and return the files (which means) we wouldn't test a final product until the final weeks of our cycle." Humongous conducts black-box testing using scripts where testers look at the game, not the code. Testers play the game and try to find typical media bugs. Humongous has localized so many games that most bugs are media-based. "As a result, testers are free to look for missing or damaged talkies, missing art or sound effects, and dialogue box problems." The level of quality rises accordingly with each subsequent localization. | 
