Methods and systems for leak detection are provided herein. A method for leak detection can comprise conveying an acoustic leak detection tool inside the innermost tubular of the multiple nested tubulars; taking measurements of the multiple nested tubulars at multiple measurement depths with the acoustic leak detection tool; arranging the measurements into a response image; and feeding the response image to a pre-trained deep neural network (DNN) to produce a flow likelihood image, wherein the DNN comprises at least one convolutional layer, and wherein the flow likelihood image comprises a representation of one or more flow patterns in at least one annulus formed by the multiple nested tubulars.