How Long Does Ear Training Take? What the Training Studies Measured

Hours for the threshold, days for intervals, weeks for vocabulary, years for the ensemble skill.

Short answer: there is no single answer, because "ear training" is at least four different skills with four different timescales. Raw pitch discrimination moves in hours. Interval comparison moves over days, and only if the sessions are arranged in a particular way. Naming vocabulary moves over weeks. Sight singing in an ensemble moves over years, and the strongest single predictor of it in the research is simply how many years someone has been doing it. Anyone quoting one number for all of that is selling something. Below is what the training studies actually measured, with the doses they used.

The measured timelines

Hours, for the threshold

The cleanest number in the field comes from Micheyl, Delhommeau, Perrot and Oxenham. They measured pitch discrimination thresholds in thirty classical musicians and thirty non-musicians, none with prior psychoacoustic training. The non-musicians started more than six times worse than the musicians. After two hours of training they were still about four times worse. Between four and eight hours of training brought them to thresholds as small as the musicians'.

Two details are easy to miss and both matter. First, the musicians barely improved, which suggests that classical training had already taken them to something close to the ceiling for this task. Second, the eight hours is laboratory time on one narrow task, not eight hours of general musicianship. What it buys you is a sharper ruler, and that is genuinely valuable, but it is not the same as knowing what to do with the ruler.

Days, and a minimum daily dose

Wright and Sabin asked a question that almost nobody asks: how much practice per day is enough for learning to accumulate across days? They trained separate groups for either 360 or 900 trials per day over six days on two auditory tasks. On frequency discrimination, more than 360 trials per day were needed for improvement to appear at all. On temporal-interval discrimination, 360 was sufficient and the extra practice added nothing.

Their interpretation is that daily consolidation only happens once a task-specific initiation requirement is met, and that training past that point yields no further gain on the trained condition. For a learner this cuts both ways. A three-minute session may genuinely be below threshold and produce nothing. A ninety-minute session is not proportionally better than a twenty-minute one, and may be no better at all.

The scheduling result that changes the arithmetic

The most surprising finding in this literature concerns musical intervals specifically. Little, Cheng and Wright compared two four-day regimens with equal numbers of stimulus presentations per day. Continuous task practice produced no learning. A regimen that alternated task practice with periods of stimulus exposure alone, where listeners simply heard the sounds without responding, produced clear improvement. Control groups given only the practice, or only the exposure, did not learn.

The improvement also transferred to an untrained stimulus type and to an untrained interval identification task, which is unusual: perceptual learning is often stubbornly specific to whatever was trained. So the honest answer to "how long does interval training take" is that it depends on whether the sessions include passive listening at all, and a regimen without it may take forever, because in this experiment it produced nothing in four days.

Years, for the ensemble skill

Sight singing is where the timescale changes character. Demorest and May tested 414 high school choir members individually and ran a regression on their musical background. The strongest predictor of individual success was years of school choir experience, followed by years of piano, then instrumental, then vocal lessons. Not method, not a training app, not a summer of intensive work: years.

That is not a counsel of despair. It is a statement about what kind of quantity you are accumulating. A threshold is a parameter you can move in an afternoon. Fluent notation reading with a voice attached is a body of practice, and it behaves like one.

The ceiling question nobody wants to hear

Practice matters, and it matters less than the popular version claims. Macnamara, Hambrick and Oswald's meta-analysis across every domain where deliberate practice had been studied found that it explained 21 per cent of the variance in performance in music. The remaining variance is starting point, age of onset, working memory, instruction quality, motivation and everything else.

Van Hedger and colleagues found something similar inside the narrow case of adult absolute pitch learning: gains were real but varied widely between individuals, and auditory working memory predicted who improved. If you train and your neighbour improves faster, that is an expected result rather than evidence that you did it wrong.

A realistic schedule

• Ten to twenty minutes a day, daily. Long enough to clear the daily minimum, short enough to sustain. Extra length is the least valuable thing you can add.
• Include passive listening. Alternate active trials with blocks of just hearing the material. On the interval evidence this is not optional padding; it may be the ingredient that makes the rest work.
• Expect the threshold first. Pitch resolution is the fastest-moving component. Weeks one and two are where you should see a number change.
• Expect vocabulary next. Naming intervals and degrees reliably is a matter of weeks of daily contact, not of hours.
• Expect the ensemble skill to take years, and stop treating that as failure. It is what the observational data says about everybody.

What our games do here, honestly

A Hearkle round takes a couple of minutes, which is well below the daily doses used in the studies above. That is a deliberate trade: a game you actually open every day beats an optimal protocol you abandon in week two. But we should be plain that a daily round is a measurement more than it is a training programme, and that a game score is not a laboratory threshold.

The useful thing a daily score gives you is a trend. Hearfork narrows a pitch difference until you stop resolving it, so its score is roughly a resolution measure and it is the one most likely to move in the first fortnight. Hearspan tracks interval reproduction. Hearecho tracks pitch memory. All eleven games are monophonic, so none of them touches chords, keys or notation, and the sight-singing timeline above is not something we can shorten for you.

Keep reading

• Interval recognition training: the session design behind the four-day interval result.
• Pitch JND: the threshold that moves fastest, and what the numbers mean.
• Can adults learn perfect pitch?: the training evidence on the hardest case.
• A four-week pitch plan: a concrete schedule built on these constraints.

Sources: Micheyl, Delhommeau, Perrot and Oxenham (2006), Influence of musical and psychoacoustical training on pitch discrimination, Hearing Research 219(1-2); Wright and Sabin (2007), Perceptual learning: how much daily training is enough?, Experimental Brain Research 180(4); Little, Cheng and Wright (2019), Inducing musical-interval learning by combining task practice with periods of stimulus exposure alone, Attention, Perception & Psychophysics 81(1); Van Hedger, Heald, Koch and Nusbaum (2015), Auditory working memory predicts individual differences in absolute pitch learning, Cognition 140; University of Chicago, Acquiring perfect pitch may be possible for some adults; Demorest and May (1995), Sight-singing instruction in the choral ensemble: factors related to individual performance, Journal of Research in Music Education 43(2); Macnamara, Hambrick and Oswald (2014), Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis, Psychological Science 25(8); McDermott, Keebler, Micheyl and Oxenham (2010), Musical intervals and relative pitch: frequency resolution, not interval resolution, is special, Journal of the Acoustical Society of America 128(4).