Speech-To-Text (STT) based mobile apps can aid Citizen Reporting Program (CRP) to reduce reckless driving in resource-constrained situations
preprintposted on 04.08.2020, 01:41 by Bhanu Pratap SinghBhanu Pratap Singh, Nirvisha SinghNirvisha Singh
Most of the data gathering devices used for monitoring driver’s behavior require large storage, strong cellular signals, and unlimited internet. Touching mobile devices, during driving, is prohibited by many law enforcement agencies. There are situations, especially in developing countries, where people get stuck on roads with a low battery, low device-memory, and no mobile network. The drivers in such situations are not able to report against reckless drivers effectively. This paper proposes the framework of the “citizen reporting program” (CRP) aided with mobile apps to reduce reckless driving in such resource-constrained situations (RCS). A mobile app was designed, developed, and tested as a tool for this purpose. It could convert speech to text without a cellular network, capture the nearest geolocation, and send data to a server on the network or internet availability. We tested its reliability for converting speech to text and got a “word error rate” (WER) of less than 5%. We tested its functional usability and got a score of more than 71% on the system usability scale (SUS). The survey showed a favorable response of 70 plus % in reducing reckless driving via CRP in RCS if aided with mobile apps.