Nigeria alone has a population of approximately 200 million citizens with an adult literacy rate of 59.6%. Meaning over 80 million adults are not fluent in English. Almost every Nigerian understands and can speak Nigerian pidgin. Nigerian Pidgin is also spoken across Equatorial Guinea, Ghana, Cameroon and of-course Nigeria. Mobile phone penetration is over 84% - just saying. I have built a prototype algorithm that at this moment does okay in recognizing pidgin, interpreting it to English and can communicate back(text to speech) in Nigerian pidgin. It's been tested by few of my colleagues and friends... And does fairly good in recognizing most of what they say in pidgin, however not as good as Siri or Alexa yet. I was wondering beyond research optimizations, if there is a business model or any commercial viability into this algorithm? Thanks!
A few things that you may want to consider to help you move forward with this:
1. Define your target market in more detail:
a. Nigerian Pidgin speakers vs. English speakers vs. both - and why
b. What platform would this be used on? (e.g. If smartphones: Define the percentage of people in Nigeria that use smartphones (e.g. 13%, 25 million).)
c. If non-English speaking Nigerian Pidgin speakers: Define the percentage of smartphone users within this group.
d. Why will these speakers use this product and how will they benefit from it?
2. What dialect is it based on and how many errors will come up with other dialects? Since there are multiple Nigerian Pidgin dialects and often ethnic groups add additional words, it may help to map out the differences to get a complete picture to help you decide on the best location and way to implement.
3. What are the potential revenue streams with this product? (e.g. product purchase for personal use vs. company use, ads, etc.)