Israeli researchers have developed an electronic ear to coach vibrato technique. Until now, the quality of a vibrato — the pulsating change of pitch in a singer’s voice — could only be judged by voice experts. Now, a Tel Aviv University research team ‘has successfully managed to train a computer to rate vibrato quality, and has created an application based on biofeedback to help singers improve their technique.’ Interestingly, this research could be used for other applications, such as improving automated help centers, where computers could be trained ‘to recognize a range of different emotions, such as anger and nervousness.’ But read more…
The three researchers work at the Department of Communication Disorders of the Sackler Faculty of Medicine of Tel Aviv University. Noam Amir and Ofer Amir are lecturers in this department, while the third scientist, Orit Michaeli, seems to only have a web page on Facebook — but I might be wrong.
Before going further, what is vibrato? It is “a musical effect than can be used when a musician sings or plays an instrument. It adds expression to a song and is created by a steady pulsating change of pitch, characterized by the amount of variation and the speed at which the pitch is varied. TAU’s application can teach singers how to mimic the vibrato qualities most attractive to the human ear.”
And here are some quotes from Noam Amir. “‘Vibrato is just one aspect of a singer’s impact,’ says Amir, an expert in the ways that emotions impact speech. ‘Singers need to arouse an emotional response, and that is a complicated task.’ […] New vocal students usually don’t have good control of their vibrato, explains Amir. ‘Their vibrato is erratic and hard to judge subjectively, and it’s hard to find to a precise measure for this. We wanted to find a way to emulate a human expert in a computer program.’”
So what did Amir and his colleagues do? “Amir’s team input into their computer many recordings by students singing vibrato and had their vibrato judged by human teachers. Using hundreds of vocal students and expert judges, the team was able to use mathematical measurements to correlate vibrato styles to their quality as judged by the teachers. The computer was then able to rate the vibrato quality of new voices on its own, producing ratings similar to those given by the expert vocal teachers. In effect, a machine had ‘learned’ how to judge the quality of an individual singer’s vibrato. The researchers then added a biofeedback loop and a monitor so that singers could see and augment their vibrato in real time.”
According to the American Friends of Tel Aviv University, the researchers demonstrated their electronic ear at the International Cultural and Academic Meeting of Engineering Students (ICAMES) held in May 2008 in Istanbul, Turkey. The news release said the team won a first price there — even if I’m unable to find a confirmation on the ICAMES website.
The original research was published in the journal Biomedical Signal Processing and Control under the title “Acoustic and perceptual assessment of vibrato quality of singing students” (Volume 1, Issue 2, Pages 144-150, April 2006).
Here is the beginning of the abstract. “While most studies that attempted to evaluate vibrato quality examined vocal productions of accomplished singers, very little is known about the characteristics of vibrato among singing students. Therefore, in this study, we performed a preliminary attempt to assess vibrato quality in their production of sustained notes. To that end, the presence and quality of vibrato in 253 sung notes was rated subjectively by five experienced singing teachers. The pitch contour was calculated for each recording, from which we calculated the FFT and the autocorrelation of this contour. Subsequently, a series of features was extracted from these two, and then different statistical methods were applied to examine whether the acoustic features could be used to define predictors that would be in agreement with the perceptual judgments. Given the moderate agreement obtained among judges, these acoustic predictors performed relatively well: vibrato existence was predicted correctly in over 82% of the recordings.”
On a related subject, you also might want to read a paper published by the same three researchers in the Journal of Voice under the title “Evaluating the Influence of Warmup on Singing Voice Quality Using Acoustic Measures” (Volume 19, Issue 2, Pages 252-260, June 2005).
Here is the beginning of the abstract. “Vocal warmup is generally accepted as vital for singing performance. However, only a limited number of studies have evaluated this effect quantitatively. In this study, we evaluated the effect of vocal warmup on voice production, among young female singers, using a set of acoustic parameters. Warmup reduced frequency-perturbation (p < 0.001) and amplitude-perturbation values (p < 0.05). In addition, warmup increased singer's formant amplitude (p < 0.05) and improved noise-to-harmonic ratio (p < 0.05). Tone-matching accuracy, however, was not affected by warmup. The effect of vocal warmup on frequency-perturbation parameters was more evident among mezzo-soprano singers than it was among soprano singers." Interesting observation...
The three researchers work at the Department of Communication Disorders of the Sackler Faculty of Medicine of Tel Aviv University. Noam Amir and Ofer Amir are lecturers in this department, while the third scientist, Orit Michaeli, seems to only have a web page on Facebook — but I might be wrong.
Before going further, what is vibrato? It is “a musical effect than can be used when a musician sings or plays an instrument. It adds expression to a song and is created by a steady pulsating change of pitch, characterized by the amount of variation and the speed at which the pitch is varied. TAU’s application can teach singers how to mimic the vibrato qualities most attractive to the human ear.”
And here are some quotes from Noam Amir. “‘Vibrato is just one aspect of a singer’s impact,’ says Amir, an expert in the ways that emotions impact speech. ‘Singers need to arouse an emotional response, and that is a complicated task.’ […] New vocal students usually don’t have good control of their vibrato, explains Amir. ‘Their vibrato is erratic and hard to judge subjectively, and it’s hard to find to a precise measure for this. We wanted to find a way to emulate a human expert in a computer program.’”
So what did Amir and his colleagues do? “Amir’s team input into their computer many recordings by students singing vibrato and had their vibrato judged by human teachers. Using hundreds of vocal students and expert judges, the team was able to use mathematical measurements to correlate vibrato styles to their quality as judged by the teachers. The computer was then able to rate the vibrato quality of new voices on its own, producing ratings similar to those given by the expert vocal teachers. In effect, a machine had ‘learned’ how to judge the quality of an individual singer’s vibrato. The researchers then added a biofeedback loop and a monitor so that singers could see and augment their vibrato in real time.”
According to the American Friends of Tel Aviv University, the researchers demonstrated their electronic ear at the International Cultural and Academic Meeting of Engineering Students (ICAMES) held in May 2008 in Istanbul, Turkey. The news release said the team won a first price there — even if I’m unable to find a confirmation on the ICAMES website.
The original research was published in the journal Biomedical Signal Processing and Control under the title “Acoustic and perceptual assessment of vibrato quality of singing students” (Volume 1, Issue 2, Pages 144-150, April 2006).
Here is the beginning of the abstract. “While most studies that attempted to evaluate vibrato quality examined vocal productions of accomplished singers, very little is known about the characteristics of vibrato among singing students. Therefore, in this study, we performed a preliminary attempt to assess vibrato quality in their production of sustained notes. To that end, the presence and quality of vibrato in 253 sung notes was rated subjectively by five experienced singing teachers. The pitch contour was calculated for each recording, from which we calculated the FFT and the autocorrelation of this contour. Subsequently, a series of features was extracted from these two, and then different statistical methods were applied to examine whether the acoustic features could be used to define predictors that would be in agreement with the perceptual judgments. Given the moderate agreement obtained among judges, these acoustic predictors performed relatively well: vibrato existence was predicted correctly in over 82% of the recordings.”
On a related subject, you also might want to read a paper published by the same three researchers in the Journal of Voice under the title “Evaluating the Influence of Warmup on Singing Voice Quality Using Acoustic Measures” (Volume 19, Issue 2, Pages 252-260, June 2005).
Here is the beginning of the abstract. “Vocal warmup is generally accepted as vital for singing performance. However, only a limited number of studies have evaluated this effect quantitatively. In this study, we evaluated the effect of vocal warmup on voice production, among young female singers, using a set of acoustic parameters. Warmup reduced frequency-perturbation (p < 0.001) and amplitude-perturbation values (p < 0.05). In addition, warmup increased singer's formant amplitude (p < 0.05) and improved noise-to-harmonic ratio (p < 0.05). Tone-matching accuracy, however, was not affected by warmup. The effect of vocal warmup on frequency-perturbation parameters was more evident among mezzo-soprano singers than it was among soprano singers." Interesting observation...
Posts
No comments:
Post a Comment