Relationships Between Intelligibility, Speaker Identification, and the Detection of Dramatized Urgency

摘要

The systems used for public safety speech communications must be intelligible. It is also desirable that they transmit secondary information, such as the attributes of a speaker's voice. This secondary information can allow a user to identify the speaker and his or her emotional state. Testing speech communications systems for the delivery of intelligible speech is common. Testing for human perception of the delivery of this secondary information is less common, though some prior work has been done. Building on this prior work, we describe a set of controlled laboratory listening experiments. These experiments characterize the relationships between speech intelligibility, speaker identification, and the detection of dramatized urgency in a speaker's voice across a range of simulated speech processing conditions. The experiment results indicate that for the speech processing conditions considered here, detection of dramatized urgency is the most robust property, speaker identification is less robust, and speech intelligibility is the least robust.