Armchair singers get ready for a computer singing coach.
A panel of X Factor judges are not needed here. A group from Tel Aviv University (TAU) has succeeded in developing computer software that can rate vibrato quality – that is, the wobbly quality a singer has while trying to sustain a note.
Showcased at the International Cultural and Academic Meeting of Engineering Students in Istanbul where it won first prize, this software application uses ‘biofeedback’ to help singers improve their technique.
Vibrato, to TAU, is something to control. Their application can teach singers how to mimic the vibrato qualities that are attractive to the human ear. New vocal students usually don’t have good control of their vibrato, explained researcher Noam Amir, a senior lecturer from the Department of Communication Disorders at the Sackler Faculty of Medicine, TAU, and an expert in the ways that emotions impact speech.
He said: “Their vibrato is erratic and hard to judge subjectively, and it’s hard to find a precise measure for this. We wanted to find a way to emulate a human expert in a computer program.”
Amir’s team entered many recordings of students singing vibrato and then had their vibrato judged by human teachers. They used hundreds of students and experts so the team could use mathematical measurements to compare the styles to the quality judged by the teachers.
Once the computer was able to judge vibrato styles on its own, the machine had therefore ‘learned’ how to judge someone’s singing. The researchers then added a biofeedback loop and a monitor so that singers could see and augment their vibrato in real time.
The original research was published in the journal Biomedical Signal Processing and Control.
Meet our Role Model who is also involved with computers and biomechanics.