Transcribed digit recognition is a critical issue in model affirmation applications, which perform computerized acknowledgement of patterns in data. Transcribed digit recognition, a subset of character acknowledgement, is the machine’s ability to perceive manually written digits. The manually written digits are of various sizes, widths, directions, and so on this, and the overall impediment is recognizing similarities in the digits and labelling them. The machine needs to confront a lot of misfortune because handwritten digits are not that good and can be made in countless unmistakable different ways. Physically composed digit affirmation is the plan for this issue since it utilizes the picture of a digit present in the training data and perceives the digit present in the picture of test data. Physically composed digit affirmation accepts a basic capacity in various client confirmation applications, for example, postal message arranging, bank check handling, structure information section, and so on. The essential objective of this digit affirmation is to ensure a strong and trustworthy philosophy for the affirmation of physically composed digits. This translated digit affirmation system isolates digits from 0 to 9. For this, Keras library in Python is utilized for characterization of MNIST dataset, an enormous information base of transcribed digits. This finding will carefully enhance the overall performance of the chirographic digital recognition with the help of quantum machine learning model.