Slangcleanser to cure Slangmania

Challenge: Understand, decode and dissuade the prevalence of ‘Slangmania’; inhibitor to the catharsis of individual and social consciousness (Indian scenario)

·    Solution:  Build an app, named Slangcleanser, which captures slang usage and give discount coupons to people who uses less slang (based on threshold) for example if one abstains from using slangs every day for 5days continuously, then he/she gets a movie voucher. Taking analogy from the elephant rider problem, this solution uses incentive to motivate the elephant. The assumption is that the rider already is aware that slang usage is bad. The point to note is that the app will be free to download, and the revenue will be generated from advertisements.

·    Before & After Storyboard: The story board depicts the intergenerational impact of slang usage. It targets young parents as customer who knows that slang is bad but by habit false prey to slang. The solution, Slangcleanser is here to address the problem by motivating them not to use slangs with coupons.

·       Looks Like Prototype: Smartphone Application Wireframe; UI Prototype

The wireframe demonstrates the various screens of the application that one user will go through. First, the app challenges the user (screen 1). Then a user is free to choose his level of challenge (screen 2). Then the app confirms the selection and tracks continuous progress (screen 3). One confident of their win can pre-book the benefits (screen 4)i.e. just to incentivise the user but if they fail to complete the challenge successfully then the benefit will get reverted. After completing the challenge successfully, the user can claim their prize,(screen 5) share their result and log out or move onto the next challenge(screen 6).            

                                       Screen 1                                                                                Screen 2

                                     

                                          Screen 3                                                                                Screen 4

                                      

                                        Screen 5                                                                               Screen 6

·       1 Day Experiment: To test the technical feasibility

The experiment is to build a prototype for the application named Slangcleanser in Android platform using free app developer platform in one day. The objective is to test whether it is possible to detect slang in conversation. For this test, a slang dictionary of 10 English slangs will be created manually. The app will access the microphone of the smartphone, record conversation and transcribe it to text. Then it will search if those slang words feature in the transcribed text. After that, it will count the number of matched slangs and display the result.

There are apps already which access microphone like TrueCaller. Also, there are apps which convert voice to text like Speechnotes. And search-finding particular words in a text is a very basic coding which is prevalent in chatting apps like WhatsApp. So, adding the bits together the technical feasibility of the idea of Slangcleanser is proved. But these application developers would not let me use their code. So, it needs to be ascertained whether I have the capability or not for this development,

·      Experiment Boundary: I will devote one day of my time. I will first make a list of 10 English slangs. I will use a free app developer software like Google app developer and try to build the android application prototype which can access microphone of smartphone, record conversation, convert it to text and compare with the list and display the count of matches.

·    Experiment Hypothesis: It is possible to detect slang in conversation in English language in android application.

·       Cost of the Experiment: There is no cost involved apart from man-hours.

·      Experiment Success Criteria:  After building the app, I will test the app. I will request my friends to try the beta version. We will converse for 10 mins. And I will request them to say some of the slangs from the list during their conversation. If the app is able to detect slangs of the list and display count, then the experiment will be termed as success.

Going Forward: If the experiment is successful then the logical next step will be to build a multi-lingual slang dictionary and test whether different slangs in different languages like Hindi, Bengali can be detected.  Jargon detection using natural language processing will have to be done also. This will establish the technical feasibility of the solution. Other dimensions of the solution like need, production and commercial will then need to be established one by one.

If the experiment is not successful, then other solutions must be tried.