The need to make visual content accessible to individuals with visual impairments has become increasingly important. Automatic image captioning has emerged as a challenging yet pivotal task in this environment. In the present study, we use deep learning architecture which is Recurrent Neural Networks (RNNs) to use the pre-trained models and integrate with Natural Language Processing (NLP) to generate descriptive captions for images. Through the extensive experimentation we used the standard dataset Flickr 8k. Our approach not only achieves the state of captions but also aims to transform the textual description of images into High quality speech which enables blind individuals to understand through auditory perception.