Developing a System to Convert Video and Audio Files to Text from Envelope Pages with Image Processing

Abstract

The objective of this study was 1) to design and develop, 2) to find the efficiency of the image and audio conversions to the envelope page, and 3) to investigate the satisfaction of the envelope scanner and the efficiency of the Image processing machine. This research was developed with a 30-sample size. The statistical method was descriptive, e.g., percentage, mean and standard deviation. The result showed that the system was efficiently tested in Thai and English. For image conversions in the Thai language, the accuracy was 35%, with an error of 65%, and English language accuracy was 91.95%, with an 8.05% error. Testing of audio conversion showed that in Thai and English, the accuracy was 91.06%, and the error in Thai and English was 8.12%. The test results showed that the accuracy of the photos might be less than the audio, which was obtained by the random character checking with voices. It was for correcting and developing the system to work with all functions. For the image and audio conversions ‘satisfaction was at a good level. ( =4.09, S.D.=1.06)