After brainstorming on problems in the world related to our shared interest in bilingualism, we decided we wanted to improve communication among people coming from different backgrounds this led us to our original question: How can we tackle the limited effectiveness of many current translations?

Your Feedback
We pitched this problem to our peers to ask them if this was Moonshot material; we wanted to make sure we were thinking big. Responses to whether it was moonshot material were a mix of green lights to go ahead and yellow lights to reconsider some specifics.

Those who said it was Moonshot material stated that it was a good starting point, and that it was a valid problem for many people, even in our class.

Those who felt that we should think a little bigger said that the problem was only vaguely stated, and that more clarification was needed on our plan to approach the problem. Some suggestions were given on narrowing the scope of the problem since there are so many different languages, we can't possibly tackle the translation issues in all of them.

Based on this feedback, we came up with...

Our New Question
How can we better convey information from a target language in hopes of modulating the tension between translation and interpretation?

We took the feedback given on specifying further what was our problem into account when revising our question. We felt the feedback on the number of different languages, on the other hand, was off-target. Our goal is to create a framework to make this possible; the specific languages come later!

Our Ideas
We took the spirit of Moonshot to heart, and started brainstorming and evaluating every idea which came to mind with Rapid Evaluation. Here are examples of some of the ideas we ended up rejecting:

  1. Improve bilingual education
    One of our first thoughts was if we want to improve cross-language communication, why not eliminate the problem at the source and have people learn more languages! Second language education is particularly lackluster in the US. However, we didn’t push forward because even in the best case scenario our problem would still exist; there are thousands of languages in the world, and no one can learn them all.
  2. Hire native translators
    Another idea was to increase the hiring of translators whose native language is that which you are translating into. Native speakers of a language tend to be better at understanding and implementing emotional cues within the language, and studies have shown the regions of the brain corresponding to emotion are more active when speaking one’s native acquired language. We scrapped this idea in the end because we realized that increasing the number of translators of a specific background is a difficult, long-term endeavor with no clear-cut manners to approach it; there were probably more effective solutions we could implement more quickly.
  3. BRAIN IMPLANTS!
    When brainstorming, we didn’t know whether or not to entertain the idea of brain implants. We began by questioning what the brain implants would allow the patient to do, but soon the question of “how” the technology would work surfaced. We concluded that, at the current time, brain implants put the patient at risk due to how invasive the operation would need to be. Moreover, the operation would be too expensive for the majority of the population to proceed with. Also, we felt as though there isn’t sufficient information or research associated with how changes in language activated regions in the brain affect target language comprehension. Therefore, it would be best to not put the lives of others at risk before there is more research conducted.
  4. An app which can add cultural subtext to translations
    The idea of creating/designing a translator became an obvious choice for our group from early on. We believed that one of the primary obstacles when translating ideas to another language revolved around cultural differences and contexts that are not universally known. We brainstormed the idea of possibly adding cultural subtexts to add more context to a translation in hopes of preventing miscommunications or ambiguities. We ended up not moving forward with this idea because we believed that it would be a time consuming task to program the translator to learn and recognize a great deal of diverse, cultural information. Traditions and customs also tend to change, evolve or even fade out as time passes so it would be difficult to account for all these changes without the proper time and effort. Since cultural contexts can often be personal, we concluded that a possible risk of improper or insufficient representation exists.
  5. Body language translator
    Once again, we brainstormed a possible feature that we believed seemed to lack in pre-existing language translators. This idea was mainly aimed towards easing misunderstandings and miscommunications during emotionally charged conversations. In this case, we thought that the issue would be resolved if we designed an app that could recognize changes in connotation based on body language and/or facial recognition. We decided not to move forward with this idea because we felt that the association between body language and implied connotation is not 100% directly correlated as we had assumed. In other words, people tend to express their inner emotions through external manners to different degrees. Also, we discussed how emotionally charged conversations/dialogues tend to be influenced by other variables such as tone syllable/word stress, or perceived loudness. The translator would have to cover many different influencing variables in conversational speech perception in order to be helpful. We felt that it would be better to single out one or two variables that can be enhanced to produce better outcomes.
  6. Context-dependent bilingual translating device
    Many words have multiple meanings, so it is often necessary to translate the entire passage rather than a single word. A translating device that takes into account the entire sentence when translating a particular word would allow one to ascertain the exact meaning of the word in that specific context, therefore eliminating confusion. We  decided not to move forward with this idea because many similar devices actually exist. Google translate generates translations for long phrases, also it generates different translating options when inputting a single word. Furthermore, many dictionaries have features that allow one to see how the particular word is used in the sentence. Because of redundancy, we decided not to move forward with this idea.

Looking ahead... which ideas do we want to move forward with?
As much fun as we had talking about every crazy idea which came forward, we eventually settled on two which we felt had the greatest potential.

  1. Prosody-preserving language translator
    We thought this idea was extremely promising. This is definitely lacking in most AI/neural network translations such as Google Translate, which focus on literal translations but fail to note the tone of voice. We read that researchers at Amazon Alexa have been working on this in the context of making their devices better at responding to voiced commands, but we think this could also be transferred to translation as well, with huge implications. Voiced translations would contain a new layer of meaning, and be more easily understand than with the bland robot voice of current translators.
  2. Audience-specific translation
    We decided to also shift our focus to the most important aspect of translation: the audience. It’s apparent that different audiences have a different breadth of everyday words and phrases that they use. Age, sex, background, geographic location can all become factors in shaping our language. That is why we came up with an audience-specific translation-- a translation device that learns the audience’s words and mannerisms (much like the predictive keyboard feature on the iPhone) to deliver more accurate translations. Since this was a promising idea, we decided to move forward with it.

We're really excited to be moving forward with these ideas to the next stage of the Moonshot framework: Rapid Evaluation Stage 2!