Building A Voice Based Image Caption Generator with Deep Learning

Image processing is used in various industries and it is remaining as one of the most advanced technologies used in Google, medical field etc. Recently, this technology has also attracted many programmers and developers due to its free and open source tool, which every developer can afford it. Image processing also helps in finding out lot of information from a single image since it is currently utilized as a primary method for collecting the information from image and processing it for some purpose and some operations will also be performed on the image. A voice based image caption generation is a task that involves the NLP (natural language processing) concept for understanding the description of an image. The combination of CNN and LSTM is considered as the best solution for this project; the main target of the proposed research work is to obtain the perfect caption for an image. After obtaining the description, it will be converted into text and the text into a voice. Image description is a best solution used for a visually impaired people who are unable to comprehend visuals. With the use of a voice based image caption generator, the descriptions can be obtained as a voice output, if their vision can't be resorted. In future, image processing will emerge as a significant research topic, which will be primarily utilized to save human lives.