According to the National Federation of the Blind, the estimated number of blind individuals in the U.S. is 1.3 million, and each year 75,000 people lose all or part of their vision. The ability to read and write is key to information access, self-sufficiency, and stronger workforce, thus directly impacts the quality of life of many visually impaired people. Braille alphabet was developed for this need in 1821 by Louis Braille, and today, it is widely and strongly accepted that study of Braille reading and writing, and development of reading and writing devices are of great importance since they play crucial role in increasing the Braille literacy. However, only less than 10 percent of the 1.3 million blind people in the US are Braille literate. This points out an important crisis in the US for the education of the blind. For broadening Braille literacy, creating new Braille reading/writing devices is therefore a critical mission. There still exist openings in Braille reading device developments, where both portability, and effective Braille reading should be addressed. Although single-cell displays promise a portable and inexpensive alternative to the existing Braille displays, they are currently not being researched due to neural adaptation problems underneath the finger. In this research project, to overcome the issues with the single-cell Braille displays, we propose a human-machine interface for the blind which extends the conventional single-cell approach to the reading process in which machine adapts to human finger speed in real-time, while displaying Braille with only few Braille cells.