Delivery of short audio communications to remote end user devices uses discrete message-oriented service messages for delivery of the audio messages. Audio information is captured, encoded by means of audio compression technology, and stored at a message center. Alternatively, the audio information may be stored in an uncompressed format and compressed immediately before delivery. The compressed audio information is transmitted by means of Short Message Service or another discrete message-oriented service to the end user device, where it is decompressed and played to the recipient. Message delivery can be accomplished in parallel with some other ongoing activity at the end user device. The major fields in a Short Message Service message used to carry encoded audio information are a Type of Data field, including information on the type of translator needed to decompress the message field, a Short Message Identifier, a Length field, and the coded audio information. If the audio information is too large to be contained in a single Short Message, segmentation and reassembly techniques commonly known in the art are applied to break the information into segments, each of which is placed into a separate Short Message having the fields: Type of Data; Short Message Identifier; Total Segments; Number of Enclosed Segment; Segment Length; and Coded Audio Information Segment. At the end user device, the Short Message can be stored until the end user listens to the audio information. The intervening network can be a radio link, the Internet, or some other network. The invention may also be utilized by a mobile user to send short audio messages to a message center and/or a second end user device.