Typewriter tinnitus is a rare, staccato tinnitus — intermittent bursts like Morse code or popcorn, typically in one ear — thought to come from a blood vessel pressing on the hearing nerve, and most patients respond to carbamazepine.
Most tinnitus is continuous. A tone, a hiss, a buzz that is there when you wake up and there when you go to bed. Typewriter tinnitus is not that. It is staccato — intermittent bursts, rhythmic and percussive — and the difference matters, because it is one of the very few tinnitus presentations with an identified physical cause and a drug treatment that usually works.
The sound is the diagnosis
The original 2006 description followed six patients with unilateral tinnitus. What is striking in it is how uniform their language was: all six described a staccato quality, reaching independently for the same handful of comparisons — "like a typewriter in the background, pop corn, Morse code". Later reports add machine-gun.
Their tinnitus had come on suddenly, at ages ranging from 39 to 87. Five of the six had no other hearing or balance complaints at all, and their audiograms were symmetric and normal for their age. That combination — a sudden, one-sided, rhythmic, intermittent sound in someone whose hearing test looks unremarkable — is what should prompt someone to name this specifically rather than filing it under ordinary tinnitus symptoms.
If your tinnitus is a steady tone, this article is not describing you.
What causes it
The accepted mechanism is neurovascular compression: a blood vessel resting against the cochleovestibular nerve, irritating it. In the original series, vascular compression on the same side as the tinnitus was found in four of the five patients who underwent imaging.
That places typewriter tinnitus in a recognisable family. Trigeminal neuralgia, hemifacial spasm and glossopharyngeal neuralgia are all syndromes in which a vessel compresses a cranial nerve and the nerve fires in bursts — facial pain, facial twitching, throat pain. The auditory version of the same thing produces bursts of sound. A related vestibular presentation, vestibular paroxysmia, can occur alongside it.
A 2023 study added an objective marker. Comparing 18 typewriter tinnitus patients against 38 age-matched controls, the affected ears showed longer auditory brainstem response interpeak latencies (I–III) and longer wave III and V latencies than both the patient's own unaffected side and the controls, with a higher proportion showing an I–III interpeak latency of 2.3 ms or more. In other words the delay is measurable, on a test audiologists already run.
The part that is unusual for tinnitus
In the original series, all six patients had their tinnitus fully suppressed by carbamazepine. In the 2023 study, 17 of 18 responded to carbamazepine or oxcarbazepine.
Response rates like that are unusual for tinnitus. It is worth being precise about why they appear here, because the temptation to over-read them is obvious.
This is a response rate for one rare, defined syndrome with an identified mechanism — a nerve being irritated by a vessel, quieted by a drug that stabilises nerve membranes. It is not a response rate for tinnitus. The American Academy of Otolaryngology–Head and Neck Surgery's clinical practice guideline recommends against anticonvulsants for the routine treatment of persistent, bothersome tinnitus (the wider picture is in drugs for tinnitus), and that recommendation and these results do not conflict: the guideline's subject is ordinary non-specific tinnitus, and the word carrying the weight in it is routine.
Carbamazepine and oxcarbazepine are prescription medications with real side effects and interactions, and in some populations a screening test is advised before starting carbamazepine. Nothing here is a reason to seek out either one; it is a reason to describe your tinnitus precisely enough that a doctor can recognise the pattern.
Whether it comes back
Sometimes. Of the 17 patients who responded in the 2023 study, 7 relapsed after stopping the drug and 10 did not.
The relapse group differed in ways that were measurable rather than guessable: significantly longer I–III interpeak latency and wave V latency, older age, and poorer hearing. That is the beginning of a way to predict who can come off medication safely — from a group of 17, so a signal rather than a rule, but a more useful one than trial and error.
If this sounds like your tinnitus
- Describe the rhythm, not just the sound. "Intermittent bursts, like Morse code, always the same ear" is a different clinical presentation from "ringing", and it is the description that gets this considered. The distinction between this and middle ear myoclonus — which also produces rhythmic clicking, from a muscle rather than a nerve — is exactly the kind of thing an ENT can separate and a search engine cannot.
- Mention that it started suddenly and stayed one-sided, if it did. Both are part of the pattern, and one-sidedness independently warrants evaluation regardless of what turns out to be behind it.
- Do not expect this to be common. It is rare. Most people who read this description will have ordinary tinnitus and the ordinary approaches will apply. But the ones who do have it are currently being treated as though nothing can be done, which is the wrong answer for this specific group.