News

Whose voice is that?

Science  11 Mar 2016:
Vol. 351, Issue 6278, pp. 1140
DOI: 10.1126/science.351.6278.1140

You are currently viewing the summary.

View Full Text

Summary

Speaker recognition is a forensic field with a checkered history that is trying to find solid scientific ground. Past methods to compare recordings of human voices were often highly subjective and have been discredited. More objective alternatives have emerged and they are reliable when comparing standard sentences clearly spoken into a microphone, but real-world recordings are often still problematic. To improve accuracy, scientists are studying how factors like inebriation, emotional state, and recording device influence voice samples. They are also testing how well existing systems perform and developing standards for things like data selection and the presentation of results.