# Tempo

**Tempo** is the rate or pace of music, typically measured as the number of perceived beats per minute (BPM). The word is Italian for "time" and entered common use in western European music during the 18th century; most verbal tempo terms in Western classical music are also Italian<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>. In music psychology, tempo is described both as a psychological assessment of how fast a piece sounds and as the physical rate of a perceived tactus, the beat a listener entrains to<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>.

| Key fact | Detail |
|---|---|
| Definition | The rate of the beat, expressed in beats per minute (BPM) or as the beat period, the time interval between successive beats<sup>[2](https://taplab.psy.msu.edu/wp-content/uploads/2014/10/McAuley_2010.pdf)</sup> |
| Verbal terms | Largo (very slow), adagio (slow), andante (moderately slow), allegretto (moderately fast), allegro (fast), presto (very fast)<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup> |
| Metronome | Patented in France on 14 September 1815 by Johann Nepomuk Maelzel, building on Dietrich Nikolaus Winkel's compound-pendulum design<sup>[3](https://www.nature.com/articles/124430a0)</sup> |
| Mechanical range | Winkel's compound pendulum design covers 40 to 208 beats per minute with a rod under 20 cm<sup>[4](https://momh.org.uk/exhibitions/johann-nepomuk-maelzel-1772-1838/)</sup> |
| Comfortable beat rate | About 70 BPM corresponds to the average human pulse or breathing rate, a link noted in theory as far back as the Renaissance writings of Gaffurius<sup>[5](https://mtosmt.org/issues/mto.16.22.3/mto.16.22.3.declercq.php)</sup> |
| Typical range | Across datasets and genres, tempo typically ranges from about 60 to 200 BPM<sup>[6](https://arxiv.org/abs/2512.18804)</sup> |

## Notation and expression

In Western classical music, tempo is expressed in two main ways: metronome markings giving an exact number of beats per minute, and verbal instructions, usually in Italian<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>. The verbal terms are ordered relative to one another, but their exact meanings are subjective; moderate tempos such as andante correspond roughly to a moderate walking speed or heart rate<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>. Before the metronome, tempos were often inferred from musical style and dance types, whose speeds were common knowledge, so composers did not necessarily need to indicate tempo on the score<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>.

Performers also shape tempo over time. Gradual accelerations and decelerations are used for dramatic effect, and rubato, literally "robbed" time, alters the relationships among rhythmic note-values by speeding up and slowing down. Early rubato involved minute changes in rhythmic values at a constant tempo, while the later 19th-century Romantic practice, associated with virtuosi such as [Franz Liszt](https://www.edgepedia.org/franz-liszt) and Frederic Chopin, took greater liberties with the tempo itself<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>.

## The metronome

For most of Western classical music history, tempo was not specified numerically, being considered obvious from musical context<sup>[7](https://www.sciencedirect.com/science/article/pii/0022460X71906444)</sup>. The mechanical solution came from a compound pendulum: a pivoted rod with two weights, in which moving the upper weight changes the center of gravity and thus the rate of swing. The Dutch mechanic Dietrich Nikolaus Winkel realized around 1812 that such a design could give slow rates from a short rod, and with a wind-up clockwork mechanism providing an audible tick, his design covered 40 to 208 beats per minute<sup>[4](https://momh.org.uk/exhibitions/johann-nepomuk-maelzel-1772-1838/)</sup>.

Johann Nepomuk Maelzel, a German inventor and friend of Beethoven, patented the device in France on 14 September 1815 after Winkel refused to sell him the rights; the Dutch Academy later denied Maelzel's claim to invention in favor of Winkel, but Maelzel set up the first factory for its manufacture in Paris in 1816<sup>[3](https://www.nature.com/articles/124430a0)</sup><sup> • </sup><sup>[4](https://momh.org.uk/exhibitions/johann-nepomuk-maelzel-1772-1838/)</sup>. Beethoven was an early adopter: metronome marks for his first eight symphonies, "determined by the composer himself according to Maelzel's metronome," appeared in Leipzig's *Allgemeine musikalische Zeitung* on 17 December 1817<sup>[8](https://www.jstor.org/stable/10.7722/j.ctt1mmftcn)</sup>. He distinguished clock time from musical time, arguing that while metronome numbers could replace subjective tempo indications, the Italian character words should remain because they denote the spirit in which the music is conceived<sup>[9](https://posgrado.unam.mx/musica/lecturas/historiaInterpretacion/RamosViterbo/Jon-Technology.pdf)</sup>.

Many of Beethoven's marks have long been regarded as too fast. One study examined whether frictional retardation of his metronome could have caused the instrument to be set at higher-than-desired frequencies<sup>[7](https://www.sciencedirect.com/science/article/pii/0022460X71906444)</sup>. Historical attitudes toward the machine were also more skeptical than modern ones: throughout much of the 19th century the efficacy of tempo machines was seriously questioned, and metronomic practice did not become universal until some point in the 20th century<sup>[10](https://academic.oup.com/edited-volume/42059/chapter/355873649?itm_content=Oxford_Academic_Books_0&searchresult=1)</sup><sup> • </sup><sup>[8](https://www.jstor.org/stable/10.7722/j.ctt1mmftcn)</sup>.

## Tempo perception

Tempo ultimately depends on the beat, or sense of pulse, and on periodicity, the perception of beats occurring at regular intervals<sup>[1](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)</sup>. Perception is not a simple reading of BPM. Beat rate can be dissociated from perceived tempo, and judgments of speed depend on which metrical level a listener entrains to: a piece at 200 quarter-notes per minute may be felt at the half-note level, which the listener experiences as 100 BPM<sup>[11](https://pmc.ncbi.nlm.nih.gov/articles/PMC11373168/)</sup><sup> • </sup><sup>[12](https://cdn.carleton.edu/uploads/sites/721/2021/12/London-2011-Tactus-vs-Tempo.pdf)</sup>. Research on speed perception has found that popular music around 115 to 120 BPM is most often perceived as neither fast nor slow<sup>[5](https://mtosmt.org/issues/mto.16.22.3/mto.16.22.3.declercq.php)</sup>.

Judgments of expressive timing, the small deviations from a steady beat, appear to be shaped more by exposure to a musical idiom than by formal training. In a web-based listening experiment by Henkjan Honing, a professor of music cognition at the [University of Amsterdam](https://www.edgepedia.org/university-of-amsterdam), and Olivia Ladinig, listeners compared recordings of the same compositions, one tempo-transformed; expressive timing judgments followed exposure to the idiom rather than expertise level<sup>[13](https://pubmed.ncbi.nlm.nih.gov/19170488/)</sup>.

## Measuring tempo in recordings

Estimating tempo from audio is a central task in music information retrieval, and successful global tempo estimation requires a stable tempo, as commonly occurs in rock, pop, and dance music<sup>[14](https://doi.org/10.5334/tismir.43)</sup>. [Deep learning](https://www.edgepedia.org/deep-learning) models have improved the task: temporal convolutional network (TCN) models trained specifically on solo instrumental performances raised the Acc1 metric for guitar tempo estimation by 38.6 percent over a pretrained model, from 61.1 percent, though accuracy on classical piano reached only 50.9 percent<sup>[15](https://arxiv.org/abs/2504.18502)</sup>.

Historical polyphonic chamber music poses harder problems. Automated beat-detection tools, designed for monophonic audio or modern studio conditions, fail systematically on recordings of Beethoven's five piano and cello sonatas (Op. 5 Nos. 1 and 2; Op. 69; Op. 102 Nos. 1 and 2), so one study developed a manual cumulative lap-timer protocol producing bar-level BPM data with millisecond resolution, which captures expressive timing phenomena such as rubato, fermatas, accelerandi, and ritardandi that automated tools suppress or misread<sup>[16](https://arxiv.org/abs/2604.15278)</sup>.

Applied to more than one hundred movement-level recordings of those sonatas spanning 1930 to 2012, this kind of measurement shows how historical tempo indications fare against actual practice. Czerny's and Moscheles's transmitted markings are consistently and substantially exceeded by the entire recording corpus, with gaps of 15 to 39 percent across movements, largest in slow Adagio movements and smallest in fast Allegro finales. Kolisch's 1943 markings align considerably more closely with recorded practice than either earlier set. Over eight decades, the central Allegro traditions for each movement have stayed stable, not due to uniformity among performers, but because slow, mid-range, and fast traditions coexist and endure at the same time, with the mid-range one prevailing throughout<sup>[17](https://arxiv.org/abs/2604.18631)</sup>.

## References

1. [Tempo, in Music in the Social and Behavioral Sciences: An Encyclopedia (Sage)](https://sk.sagepub.com/ency/edvol/embed/music-in-the-social-and-behavioral-sciences/chpt/tempo)
2. [J. Devin McAuley (2010). Tempo and Rhythm, in Music Perception, Springer Handbook of Auditory Research](https://taplab.psy.msu.edu/wp-content/uploads/2014/10/McAuley_2010.pdf)
3. [Calendar of Patent Records, Nature 124, 430 (1929)](https://www.nature.com/articles/124430a0)
4. [Johann Nepomuk Maelzel (1772-1838), Museum of Music History](https://momh.org.uk/exhibitions/johann-nepomuk-maelzel-1772-1838/)
5. [de Clercq, Measuring a Measure, MTO 22.3](https://mtosmt.org/issues/mto.16.22.3/mto.16.22.3.declercq.php)
6. [Tempo as the Stable Cue: Hierarchical Mixture of Tempo and Beat Experts for Music to 3D Dance Generation, arXiv](https://arxiv.org/abs/2512.18804)
7. [A note on Beethoven's metronome, Journal of Sound and Vibration](https://www.sciencedirect.com/science/article/pii/0022460X71906444)
8. [Beverly Jerold (ed.), Music Performance Issues: 1600-1900, Pendragon Press](https://www.jstor.org/stable/10.7722/j.ctt1mmftcn)
9. [Technology and Music Performance in the Age of Mechanical Reproduction](https://posgrado.unam.mx/musica/lecturas/historiaInterpretacion/RamosViterbo/Jon-Technology.pdf)
10. [Alexander E. Bonus, Metronome, Oxford Handbook Topics in Music](https://academic.oup.com/edited-volume/42059/chapter/355873649?itm_content=Oxford_Academic_Books_0&searchresult=1)
11. [The influence of tonality, tempo, and musical sophistication on the listener's time-duration estimates, PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC11373168/)
12. [London, Tactus vs Tempo, Empirical Musicology Review](https://cdn.carleton.edu/uploads/sites/721/2021/12/London-2011-Tactus-vs-Tempo.pdf)
13. [Honing & Ladinig (2009). Exposure influences expressive timing judgments in music, J Exp Psychol Hum Percept Perform](https://pubmed.ncbi.nlm.nih.gov/19170488/)
14. [Schreiber et al., Music Tempo Estimation: Are We Done Yet?, TISMIR](https://doi.org/10.5334/tismir.43)
15. [He, Togneri & Zhang (2025). Music Tempo Estimation on Solo Instrumental Performance, arXiv](https://arxiv.org/abs/2504.18502)
16. [A Manual Bar-by-Bar Tempo Measurement Protocol for Polyphonic Chamber Music Recordings, arXiv](https://arxiv.org/abs/2604.15278)
17. [Towards Revised Tempo Indications for Beethoven's Piano and Cello Sonatas, arXiv](https://arxiv.org/abs/2604.18631)

---
*Topic: Encyclopedia › Arts and culture › Music › Musical practice and theory › Instruments, theory, and world traditions › Rhythm, meter, and time in music*

*Initially written Sep 17, 2026 · Reviewed: Sep 30, 2026; Oct 25, 2026 · Edited: Sep 30, 2026; Oct 25, 2026 · Last review: Oct 25, 2026*

*Copyright 2026 EdgeChat AI, a subsidiary of Biostate AI.*

License: Edgepedia Community License 1.0, https://www.edgepedia.org/edgepedia/license
