Konnakol Drum
Say the rhythm — ta ka di mi, tha dhi thom nam, dha dhin dhin dha — and hear it on the drum, with every syllable written out.
Your syllables appear here, one line per phrase. Tap any of them to correct it.
matched to your voice, typed or correcteda guess — tap to correct it
Teach it your voice
Out of the box the tool guesses each syllable from general rules about speech, and it will get some wrong. Thirty seconds of teaching fixes most of that: tap a syllable and say it three times, or let it walk you through the main six. It keeps a small fingerprint of each — on this device only, never the sound itself.
Correcting a syllable in the phrase above teaches it too — so the tool keeps improving as you use it.
Type it, tap it, or use a recording
Type a phrase the way you would write it — spaces are optional, a comma is a held beat (kārvai) — and hear it on the drum.
Have a recording of a recitation? and the drum plays along with it.
Recite a rhythm, hear it on the drum
In Carnatic music a rhythm is spoken before it is played. The spoken syllables — solkattu — each stand for a stroke on the mridangam, and performing them with the voice is an art of its own: konnakol. A student learns a korvai by reciting it, long before their hands can play it. Hindustani music does the same with tabla bols.
The Konnakol Drum closes that gap. You recite into your phone or laptop; it writes down what it heard and plays the phrase back with mridangam or tabla strokes. There is nothing to install and no account, and the sound never leaves your device.
Three ways to use it
- Practice. Recite a phrase and pause. The drum answers with the same phrase, in your own timing, while the syllables stay on screen so you can check what you said against what you meant. Tap Replay to hear everything again.
- Perform. The drum plays each stroke as you say it, so a vocalist or konnakol artist becomes their own percussionist. Expect the stroke about a tenth of a second after the syllable begins — the tool has to hear the vowel before it can tell “ta” from “thom”. Use headphones.
- Tabla. Switch the drum and recite bols instead — dha dhin dhin dha, dha dhin dhin dha, dha tin tin ta, ta dhin dhin dha is tintāl's theka.
Which syllable plays which stroke
| Mridangam | Stroke |
|---|---|
| ta · tha | Ta — a short, sharp stroke on the rim of the valantalai (treble head) |
| ka · ki | Ki — a closed, damped tap |
| di · dhi | Din — the open, ringing stroke in the middle of the valantalai |
| mi · nam · na | Nam — the bright ringing rim stroke |
| thom | Thom — the deep open bass of the thoppi |
| tham | Thom and Nam together |
| dheem | Thom and Din together |
| gu | Gumki — the bass bent upward with the heel of the hand |
| Tabla | Stroke |
|---|---|
| dha | Ghe on the bāyāñ with Na on the dāyāñ |
| dhin | Ghe with Tin |
| na · ta | Na — the ringing rim stroke of the dāyāñ |
| tin | Tin — the clear stroke just inside the rim |
| te · ti | Te — a closed slap on the dāyāñ |
| ke · ki · ka | Ke — the flat, damped bass |
| ge · ghe | Ghe — the open bass with its rising “wah” |
| tun | Tun — the deep open centre of the dāyāñ |
These are the commonest pairings. Schools and teachers differ — the same “ta” may be played with either hand depending on what surrounds it — so treat the drum's answer as a guide to the rhythm, not as fingering.
How it tells syllables apart
There is no speech model behind this, and nothing is sent to a server. The page listens the way a phonetics textbook would:
- Where a syllable starts. Two hundred times a second it measures the sound in 24 frequency bands. A sudden jump across the bands, above the hum of the room, is the start of a syllable.
- Which vowel. “Ah”, “ee” and “oh” put their energy in different places — “ee” low and high with a gap between, “ah” in the middle, “oh” low only. That separates ta from di from thom.
- How the consonant began. A hum before the vowel means a nasal — nam, mi. A puff of breath means an aspirated stop. A fainter buzz cut off by a burst means a voiced one, the dh of dha.
- What usually follows. After ta, a doubtful syllable is more likely ka than another ta.
That is enough to be useful and not enough to be perfect: the difference between ta and ka is a few hundredths of a second of noise, and every voice, accent and microphone shapes it differently. Which is why the tool can learn your voice. Once it has a few examples of how you say each syllable, it stops guessing and simply matches each new sound to the closest one you taught it.
Getting the best results
- Hold the phone about a hand's width from your mouth, in a quiet room.
- Recite clearly, each syllable separate. Start slower than you think you need to.
- Teach it your voice at the speed and pitch you normally recite at.
- If it hears syllables that are not there, lower the mic sensitivity. If it misses soft ones, raise it.
- For long phrases with held beats, lengthen the pause the drum waits for, so it does not answer mid-korvai.
What is konnakol?
The South-Indian art of performing percussion syllables with the voice. The syllables — ta, ka, di, mi, thom, nam and many more — are called solkattu, and each stands for a mridangam stroke. Students recite before they play, and konnakol is also performed on stage in its own right.
Is my voice recorded or uploaded?
No. The sound is analysed in your browser and discarded as it goes; nothing is recorded or sent anywhere. If you teach it your voice, a small set of numbers describing each syllable is kept on this device so it remembers you next time — “Forget my voice” deletes it.
Why does it write the wrong syllable?
Until it has been taught, it is guessing from general rules, and consonants like t and k are easy to confuse. Tap the wrong syllable and choose the right one — that corrects the phrase and teaches the tool at once. A dashed outline marks a guess; a solid one was matched to your voice.
Do I need headphones?
Not in Practice mode — the microphone is ignored while the drum answers. In Perform mode the drum plays while you speak, so a loudspeaker can feed the drum back into the microphone. Headphones stop that; without them, keep the volume low.
Why is there a small delay in Perform mode?
“Ta”, “thom” and “tham” all begin with the same consonant; the drum cannot know which stroke to play until it has heard a little of the vowel. That takes about a twentieth of a second, plus whatever delay your device adds to sound going in and out. Wired headphones are quicker than Bluetooth ones.
Are these real drum recordings?
No — the strokes are synthesized in your browser, using the same sound as the Online Mridangam and Online Tabla. That keeps the page small and lets you tune the drum to your own Sa.
Is it free?
Yes. No sign-up, no download, no limit.
More free music tools: Online Mridangam · Online Tabla · Metronome · Tanpura · Carnatic Ragas