Expert listeners can often form an immediate impression of a singer’s performance. But what exactly in the voice drives that judgment? In classical vocal competitions, performers are rated on an overall score. However, the criteria behind that score are rarely made explicit. Moreover, voice quality is often described in vague, metaphorical terms, such as “ringing,” “warm,” or “focused,” which can mean different things to different listeners.

To better understand what lies behind these judgments, a team of researchers led by Assistant Professor Haruka Kondo from Graduate School of Media and Governance, Keio University, Japan, along with Dr. Sotaro Kondoh from the Faculty of Environment and Information Studies, Keio University, and Japan Society for the Promotion of Science, Japan, and Associate Professor Shinya Fujii from the Faculty of Environment and Information Studies, Keio University, Japan, simulated a vocal competition in a concert hall, asking expert judges to evaluate the singers and investigating which vocal attributes and acoustic features were associated with higher overall scores. The study, to be published online in the Journal of Voice on October 27, 2026, provides measurable ways to understand how experts evaluate classical singers.

Assistant Professor Haruka Kondo is a Ph.D. student in vocal science at Keio University and an operatic soprano who has competed in various vocal competitions. She conducted the study with Dr. Sotaro Kondoh, Ph.D., a Japan Society for the Promotion of Science Research Fellow and Visiting Research Fellow in musical meter, beat, and rhythm perception, and Associate Professor Shinya Fujii, Ph.D., who specializes in the neuroscience of music and directs Keio University’s Neurosciences and Music Laboratory. The researchers are affiliated with Keio University’s Research Center for Music Science, which uses scientific approaches to study how music is perceived, understood, created, and experienced, as well as its effects on the brain, mind, and body.

“Our study bridges the gap between artistic intuition and science and does not aim to replace the expert ear. It helps translate what experienced listeners hear into a shared, measurable language,” says Ms. Kondo.

The researchers recruited 22 classical singers, including 10 men and 12 women. Each singer performed the same song, “Caro mio ben,” a cappella, without instrumental accompaniment, in a concert hall to recreate the conditions of a live performance. Five expert judges gave each performance an overall score on a 100-point scale and rated six vocal attributes on a 7-point scale, covering aspects such as pitch variation (vibrato), voice quality (resonance and timbre), clarity of words (diction), pitch accuracy (intonation), and expressiveness.

The participants’ voices were also analyzed for four acoustic features: singing power ratio, which reflects the prominence of certain frequencies in the voice; loudness units relative to full scale, a standardized measure related to perceived loudness; vibrato rate, which describes how quickly the pitch changes during vibrato; and vibrato extent, which describes the amount of pitch change during vibrato. The researchers then examined which vocal attributes and acoustic features were associated with higher overall scores and whether these relationships differed between male and female singers.

Notably, judges gave higher overall scores to singers rated as having better vibrato, resonance, and expressiveness, while timbre, diction, and intonation showed no clear link to scores. Voices with greater spectral prominence in the 2–4 kHz range also tended to receive higher scores in both men and women. This range is linked to the “singer’s formant,” which is thought to give trained voices a clear, ringing quality that helps them carry without a microphone. Higher vocal loudness, however, was not associated with higher scores.

It was observed that judges evaluated male and female singers differently based on their vibrato. Among male singers, a faster vibrato was linked to lower scores, while among female singers, a wider vibrato was linked to higher scores.

“Our results show that something as personal and artistic as ‘a beautiful voice’ can be partly understood through science. This could help students understand what their instructors mean when they say, ‘let it ring more’ or ‘your vibrato is too shallow.’ Students could then use these measurements to see how their voices change as they practice between lessons, which could help organizers design clearer scoring sheets and help singers understand their results,” says Ms. Kondo.

The researchers also see potential for these measures to improve singing-training apps and artificial intelligence voice-evaluation tools that give users feedback that reflects expert judgement.

“In summary, our study paves the way for more efficient and concrete teaching in music education, greater transparency and fairness in competitions and auditions, and designing singing evaluation tools that better reflect how experts listen,” concludes Ms. Kondo.Reference
Title of original paper: Vocal Evaluation of Classical Singing in a Concert Hall: Perceptual and Acoustic Associations with Overall Scores in Male and Female Singers
Journal: Journal of Voice
DOI: https://doi.org/10.1016/j.jvoice.2026.09.070