The audio signal obtained by a receiver from a sound source depends on the sound environment and the location of the receiver relative to the source. When an audio signal is given, it is necessary to find the best location of the receiver to obtain the audio signal. This paper presents a sound receiver location estimation method using a convolutional neural network. The sound receiver’s location estimation task is comprehended as an im-age classification problem; in which we aim to classify a given audio signal according to the location of the receiver. Rectangular audio rooms are simulated with different dimen-sions and surface materials. The audio signal obtained by a receiver from a fixed sound source in the simulation room is calculated and simulated via impulse response by the image source model. Then, the audio signals are transformed into spectrograms, allowing the convolutional layers to extract the appropriate features required for classification. After datasets are trained and tested, the proposed convolutional neural network model with op-timal hyperparameters exhibits high audio signal identification accuracies for all the sim-ulation rooms. Using the proposed model, an experiment testing the receiver’s estimated location in an experiment room was conducted, and the results indicate an identification accuracy of 97.6%. The research can also be applied to obtain optimal sound quality and design of an audio room.