A novel technique is presented to identify the codec of a coded audio. The technique does not perform decoding, utilize any coding metadata, or assume information about the structure describing the bit stream format of a codec. The underlying idea of the technique is that the design choices governing the compression level, audio quality and complexity of a codec will reveal themselves on the coded audio. To exploit this, the technique samples 2-4 kilobytes of data from a coded audio and analyzes the randomness and chaotic nature of the sampled data to build statistical models that represent encoding process associated with different codecs. In experiments, we utilize 16 of the most popular audio codecs used for high quality audio compression and in PSTNs, cellular networks, and VoIP networks by setting encoding parameters of each codec to its most commonly used values. Results show that the codec of an encoded audio can be identified with an accuracy of more than 95 percent. Other experiments considering several transcoding scenarios were also performed. Those results show that the scheme can even discriminate the first encoder of a doubly-encoded audio with an accuracy range of around 80 to 90 percent or more as long as the second codec operates on higher bit rates than the first one. (C) 2013 Elsevier Inc. All rights reserved.