Please extract all the text from the image. The output should be a JSON object with a single key "answers" which is an array of objects. Each object in the array should have two keys: "question" and "answer". The "question" key should contain the original question asked by the user. The "answer" key should contain the extracted text from the image, formatted as a JSON string.