Weekly Speaking Minutes: The Language Metric Worth Tracking
Most learners have a good idea of how much they listen and read. Ask how many minutes they actually spoke the language last week and the answer is usually a shrug, or an awkward "maybe ten?". That gap is exactly why minutes spoken per week is worth tracking. Speaking is the skill most people say they want and the one they practise least, and a weekly number makes the gap visible.
On this page
Why weekly, not daily
Speaking practice is lumpy. You might have a 60-minute tutor lesson on Tuesday, a voice message exchange on Thursday and nothing else. A daily speaking target would fail on most days and push you towards token practice. A weekly total smooths that out and matches how conversation actually fits into a life: a few sessions a week, not a few minutes every day.
What counts as speaking time
Be strict, or the number stops meaning anything:
- Conversation with a tutor or partner. Count the session, but be honest about your share. In a 60-minute lesson where the tutor explains grammar for 20 minutes, you didn't speak for 60. Many people log the full session as speaking and accept that the number is a ceiling; others estimate their share. Pick one approach and stay consistent.
- Language exchanges. Only the half in your target language.
- Self-talk and recorded monologues: describing your day aloud, retelling a video, answering practice exam questions. These count, and they're the easiest minutes to add without a partner.
- Shadowing sits in between. It trains pronunciation and rhythm but isn't producing your own sentences. Log it as speaking if you like, but in a separate source so you can tell the two apart.
What doesn't count: reading aloud from a textbook as filler, or listening to others talk in a group class while you say three sentences.
What targets to set
There's no research-backed number for "minutes of speaking per week", so treat these as starting points to adjust, not rules. They assume you're also getting plenty of input.
| Stage | Minutes a week | Typical shape |
|---|---|---|
| Beginner (A1–A2) | 30–60 | One short lesson, plus a few minutes of self-talk |
| Intermediate (B1) | 60–120 | One or two conversations, one recorded monologue |
| Upper-intermediate (B2+) | 120–180 | Two or three real conversations |
| Exam or move in 3 months | 180+ | Several sessions, including exam-format practice |
The most important number is not the target but the trend. Going from 0 to 45 minutes a week is a bigger change than going from 120 to 150. If you've never tracked it, log a normal week first and set your target at about 1.5 times that.
An Arabic example
Arabic shows why this metric needs a little thought. The FSI puts Arabic in its hardest category, at about 2,200 class hours for English speakers to reach professional working proficiency (see the FSI list), so speaking progress is slow and easy to lose sight of. It's also diglossic: Modern Standard Arabic is used for writing, news and formal speech, while everyday conversation happens in a spoken dialect such as Egyptian or Levantine.
That makes "minutes of Arabic spoken" ambiguous. A learner who reads news in MSA and talks with a tutor in Levantine should log the two separately, for example with the source or a note. A reasonable weekly picture for an intermediate learner might look like this:
- Tutor conversation in Levantine: 2 × 30 minutes
- Voice messages with a friend: about 15 minutes over the week
- Recorded two-minute retelling of a news clip in MSA: 3 × 5 minutes, including retakes
That's about 90 minutes of speaking a week, a solid intermediate target, and you can see at a glance that most of it is dialect. More on the rest of an Arabic log in tracking Arabic learning progress.
Pair the minutes with a quality check
Minutes measure effort, not ability. Once a month, record yourself talking for two minutes on a fixed prompt ("describe your week", "explain your job") and keep the recordings. Listening back to month one in month six is the clearest evidence of speaking progress you will get, far better than any app score. Some learners also count conversations per week, since starting a conversation is often harder than continuing one.
Keep speaking next to input, not instead of it
Speaking minutes are best read next to your input hours. If you're getting 7 hours of listening a week and 20 minutes of speaking, that's a typical early ratio and fine. If you're at B2 in comprehension after a year and still speaking 10 minutes a week, the log tells you where the next hour should go. Input hours vs output hours covers that balance in more detail.
Tracking it without extra work
A spreadsheet with an activity column does the job: filter for "speaking" and sum by week. In the LangTrack tracker, every entry has an activity, so tagging a session as speaking keeps it separate from your listening and reading, and a weekly goal gives you a minutes target that doesn't punish a quiet Monday. Your tutor or exchange partner can be a source, so a regular lesson becomes a one-tap shortcut. Log this week's conversations and see where you are.
FAQ
How many minutes a week should I practise speaking?
There's no research-backed figure. As a starting point, 30 to 60 minutes a week works for beginners, 60 to 120 for intermediate learners and two to three hours or more if you're preparing for an exam or a move. Log a normal week first and aim for about 1.5 times that.
Does a one-hour tutor lesson count as one hour of speaking?
It counts as one hour of lesson, but you probably spoke for less. Either log the full session and treat the speaking total as a ceiling, or estimate your share. Stay consistent so week-to-week comparisons hold.
How do I track Arabic speaking practice with MSA and dialect?
Log them separately, for example with a different source or note for Modern Standard Arabic and your dialect. Most conversation happens in dialect, so the split shows whether your speaking time matches what you'll actually use.
See your speaking minutes
Tag sessions as speaking and set a weekly goal. Free for one language.
Start tracking — free