Abstract

this paper we describe a new testbed for developing speech recognition algorithms- the ARPA-sponsored VoiceMail transcription task, analogus to other tasks such as the Switchboard, CallHome [1] and the Hub 4 tasks [2] which are currently used by speech recog- nition researchers. As the name indicates, the task involves the transcription of voicemail conversations. Voicemail represents a very large volume of real-world speech data, which is however not particularly well represented in existing databases. For instance, the Switchboard and CallHome databases contain telephone conversations between two humans, representing telephonebandwidth spontaneous speech; the Hub 4 database contains radio broadcasts which represents different kinds of speech data such as spontaneous speech from a well-trained speaker, conversations between two humans possibly over the telephone, etc. The Voicemail database on the other hand also represents telephone bandwidth spontaneous speech, however the difference with respect to the Switchboard and CallHome tasks is that the interaction is not between two humans, but rather between a human and a machine- consequently, the speech is expected to be a little more formal in its nature, without the problems of cross-talk, barge-in etc. This eliminates some of the variables and provides more controlled conditions enabling one to concentrate on the aspects of spontaneous speech and effects of the telephone channel. In this paper, we will describe the modaltry of collection of the speech data, and some algorithmic techniques that were devised based on this data. We will also describe the initial results of tran- scription performance on this task

Citations