How many hours of aphasia practice per week is enough?

The evidence points more clearly to total hours and to how many days a week you practise than to hours per week. In the RELEASE analysis of 959 people, the largest overall language gains were associated with a total of more than 20 and up to 50 hours of therapy, and with practising 3 to 5 or more days a week. Very small total amounts did not help: no gains in functional communication were seen at 5 hours or less in total.

Is it better to do one long session or several short ones?

Several short ones, on the current evidence. Frequency showed a clearer relationship with recovery than session length. Practising a little on most days is more consistent with what worked in the studies than one long weekly session.

Does practising more always produce more improvement?

No, and the best available evidence does not show a simple more-is-better pattern. In the RELEASE analysis the relationship between weekly intensity and gains was not a straight line: substantial gains appeared at several different weekly intensities. Practising a great deal on the wrong targets does not help, and fatigue reduces the quality of practice.

Does research on speech therapy apply to practice with software?

Not directly, and this is important. The dosage studies examined therapy delivered by speech-language pathologists. Applying those numbers to independent practice with software is a reasonable starting assumption, not a proven equivalence. Software practice is best understood as a way to add practice hours between sessions with a clinician, not as a substitute for them.

Is it too late if the stroke was years ago?

No. The study of Parrot Software by Corwin and colleagues enrolled people between 11 and 85 months past the onset of their aphasia, an average of about three years, and still found significant improvement in naming words that had never been practised. Chronic aphasia can still respond to treatment.

How much aphasia practice is enough?

Total hours matter more than hours per week, and practising most days beats one long session. Here is what the evidence actually shows.

By Frederick F. Weiner, Ph.D., speech-language pathologist. Former Professor in Charge of Speech Pathology, The Pennsylvania State University. Last reviewed 15 August 2026.

The short answer

The largest analysis of this question to date pooled data from 959 people across 25 randomised trials. The clearest findings were about total hours and how many days a week practice happened — not about how many hours were packed into each week.

The greatest gains in overall language were associated with more than 20 and up to 50 hours of therapy in total.
Five hours or less in total produced no gains in functional communication. Dabbling does not work.
Practising 3 to 5 or more days a week was associated with the best outcomes. For comprehension specifically, gains appeared only at 4 to 5 days a week.

Put simply: a modest amount on most days, sustained until the total reaches tens of hours, matches the evidence better than an occasional marathon.

What the studies found

Three pieces of research carry most of the weight here.

RELEASE (2022) — 959 people, 25 trials

An individual participant data network meta-analysis published in Stroke. Because it pooled data on individual people rather than study averages, it could separate total dosage, weekly intensity, and weekly frequency — which earlier work could not. Total dosage and frequency showed the clearest relationships with recovery.

Bhogal, Teasell and Speechley (2003)

Reviewed eight studies. The four in which patients improved averaged 8.8 hours of therapy per week for 11.2 weeks. The four showing no change or worse averaged 2 hours per week for 22.9 weeks. Similar total hours, spread thin, did not work.

Robey (1998)

A meta-analysis of aphasia treatment outcomes which found that more than 2 hours of treatment per week produced greater gains than less than 2 hours per week.

What this research does not tell you

We sell practice software, so we want to be careful not to claim more than the evidence supports.

These studies examined therapy delivered by clinicians, not independent practice with software. Treating the two as interchangeable is an assumption, not a finding. Practice software is a way to add hours between sessions with a speech-language pathologist — not a replacement for one.
More is not reliably better. In the RELEASE analysis the relationship between weekly intensity and improvement was not a straight line. Large gains appeared at several different weekly intensities, including relatively low ones. Anyone telling you that recovery scales smoothly with hours practised is going beyond the data.
Hours on the wrong targets are wasted. Dosage is only meaningful once the practice is aimed at the right difficulty and the right skills. That is a clinical judgement, and it is the main reason to keep a speech-language pathologist involved.
Averages are not people. These figures describe groups. Recovery varies enormously with the type and severity of aphasia, the site of the lesion, time since onset, fatigue, mood, and support at home.

A reasonable place to start

If a speech-language pathologist is involved, follow their plan — they know the targets. Failing that, a schedule consistent with the evidence above looks roughly like this:

20 to 30 minutes, 4 or 5 days a week — frequency first, length second.
Stop when fatigue sets in, not when the timer says so. Practice done while exhausted is poor practice, and in aphasia fatigue is a genuine clinical factor rather than a lack of effort.
Expect it to take months, not weeks. At half an hour five days a week, reaching 20 hours takes about two months. That is the scale the research is describing.
Plateaus are normal. Progress in aphasia is uneven. A flat few weeks is not a reason to stop.

What this means if the stroke was a long time ago

People are often told that recovery stops after six months or a year. The research on chronic aphasia does not support that.

In the study of our own software by Corwin and colleagues at Texas Tech University Health Sciences Center, participants were between 11 and 85 months past the onset of their aphasia — an average of about three years. After a 32-hour protocol over four weeks, they showed statistically significant improvement in naming words that had never been practised during treatment. Six participants is a small study, and there was no control group, but the direction is consistent with a wider literature showing that chronic aphasia remains treatable. The full text, along with its limitations, is on our research page.

Common questions

How many hours of aphasia practice per week is enough?: The evidence points more clearly to total hours and to how many days a week you practise than to hours per week. In the RELEASE analysis of 959 people, the largest overall language gains were associated with a total of more than 20 and up to 50 hours of therapy, and with practising 3 to 5 or more days a week. Very small total amounts did not help: no gains in functional communication were seen at 5 hours or less in total.
Is it better to do one long session or several short ones?: Several short ones, on the current evidence. Frequency showed a clearer relationship with recovery than session length. Practising a little on most days is more consistent with what worked in the studies than one long weekly session.
Does practising more always produce more improvement?: No, and the best available evidence does not show a simple more-is-better pattern. In the RELEASE analysis the relationship between weekly intensity and gains was not a straight line: substantial gains appeared at several different weekly intensities. Practising a great deal on the wrong targets does not help, and fatigue reduces the quality of practice.
Does research on speech therapy apply to practice with software?: Not directly, and this is important. The dosage studies examined therapy delivered by speech-language pathologists. Applying those numbers to independent practice with software is a reasonable starting assumption, not a proven equivalence. Software practice is best understood as a way to add practice hours between sessions with a clinician, not as a substitute for them.
Is it too late if the stroke was years ago?: No. The study of Parrot Software by Corwin and colleagues enrolled people between 11 and 85 months past the onset of their aphasia, an average of about three years, and still found significant improvement in naming words that had never been practised. Chronic aphasia can still respond to treatment.

References

RELEASE Collaborators. Dosage, Intensity, and Frequency of Language Therapy for Aphasia: A Systematic Review-Based, Individual Participant Data Network Meta-Analysis. Stroke. 2022; 53: 956–967. doi:10.1161/STROKEAHA.121.035216
Bhogal, S. K., Teasell, R., & Speechley, M. Intensity of aphasia therapy, impact on recovery. Stroke. 2003; 34: 987–993.
Robey, R. R. A meta-analysis of clinical outcomes in the treatment of aphasia. Journal of Speech, Language, and Hearing Research. 1998; 41: 172–187.
Corwin, M., Wells, M., Koul, R., & Dembowski, J. Computer-Assisted Anomia Treatment for Persons with Chronic Aphasia: Generalization to Untrained Words. Journal of Medical Speech-Language Pathology. 2014; 21(2): 149–163. Full text

Practising with Parrot Software

Parrot Software is a way to put more practice hours into a week between sessions with a speech-language pathologist. One subscription includes 94 programs across speech, reading, memory, vocabulary, function, and cognition, and speaking programs use speech recognition to give instant feedback. It runs in a browser on any computer, tablet, or phone, in English and Spanish.

The first week is free and does not require a credit card, so you can find out whether daily practice is realistic before paying for it. See pricing.