Your challenge is to build an app which includes the following functionalities.
- Camera access: The app must have access to the smartphone camera, allowing the user to take snapshots of any kind of text (from paper or monitor) to use it as input (see Google translate app).
- Text recognition: The app must be able to use the image from the smartphone camera as an input (Optical Character Recognition) allowing the user to select the acronym by highlighting it on the mobile screen.
- Handwriting recognition: The app must be able to use a handwritten text as input.
- Voice recognition (optional): The app should be able to use a spoken text as input.
- Acronym glossary: The app must be able to retrieve basic acronyms like (KPI, POC, etc.) and company-specific acronyms as well – English only. If possible, it could be useful if the app would be able to understand the context (correlated to camera text recognition) and retrieve the right acronym explanation.
- Database: Use/build a Nokia internal database that has all the acronyms, which can be further maintained and enriched through each users’ contributions: by adding, updating Acronym, Definition, Function/Unit etc. (grouping capabilities)
- Data maintenance/expansion: The app must contain a functionality to easily maintain and expand the existing data (e.g. Wikipedia-like input)
- Data access: The data must be also accessible via a full list of acronyms and a search which will return specific acronyms (filtering capabilities). If possible, the search should be done in multiple sources/multiple APIs.
- Offline functionality: The app must also work offline when there is no internet access.
- Operating system: The app must run on iOS and Android.
If a demo proving these capabilities (or even more) is already available, we encourage you/your team to participate and showcase. Sharing a link or any documentation helping you to prove your capabilities in this area is also appreciated.
The developed app serves as an experiment to understand the delivery feasibilities when it comes to text recognition capabilities through a mobile camera read OCR (Optical Character Recognition) & the potential to enrich a database through a Wikipedia-like input concept, where everyone can contribute. If this experiment delivers a powerful outcome, going into production could be considered as a next stage (but not part of this experiment).
The application prototype should be developed and available early December 2017. It should be self-contained, with no dependencies to Nokia environment but with the option to be connected at any point (authentication, maintenance ownership to be considered).