Klangtal is a multi-tap delay with two lanes of echoes. Each lane is a river in an alpine valley, each buoy on it is a tap, and effects sit before the delay, inside its feedback or after it.
Two minutes from opening Klangtal to echoes you have placed yourself. It runs in the browser, and the sound needs a first tap.
The valley is the instrument. The header and the tabs sit around it, and a card opens for whatever you tap.
After a few seconds without a touch the labels fade to dots, so you can see the valley. Any tap, drag or key brings them back. A tap on empty scenery while they are faded only brings them back.
It is a delay first: sound goes in, and comes back later as echoes you can see.
Two lanes, each a river. Whatever you play goes into lane A and lane B at once. Each lane is a delay line drawn as a river: sound enters at the glacier and runs downstream. The two lanes have their own taps and steps, and share their effect settings.
Taps are buoys. A tap is one point where the lane plays an echo. A lane has 1 to 8 taps, and new setups start with two. The further downstream a buoy sits, the later its echo. The whole river is as long as time (or delay beats when synced), and stretch spreads the buoys further apart or closer together.
Effects sit in three places. Before changes the sound as it enters the lane, once. Feedback is the channel that carries echoes back to the top: an effect there changes every echo again each time it comes round, so its sound builds up in the tail. After changes only what you hear, and never feeds back. The same effect sounds very different in each place. See fx.
Cross is the one bridge. Echoes coming back up the feedback channel can spill into the other river. Cross sets how much: at 26% each return is 74% its own lane and 26% the other. Zigzag has every other buoy listen to the other lane instead.
Steps let new sound in. Each lane has a row of 4 to 16 steps, two to a beat. A closed step keeps new sound out of that lane, but the echoes already in it carry on. Lanes of different lengths drift apart and meet again. See rhythm.
Klangtal is an effect, so it needs something to play into it. source in the header opens a list:
stop stops whatever is playing and clears the echoes. The source isn't an edit, so undo never changes it, and it isn't saved with a preset.
Tap a buoy, an effect, a lane badge or the cross pill to select it. Drag a buoy along its river to move its tap. Drag an effect to move it to another place or along its zone. Press near a river, away from any buoy, and the nearest buoy on that river follows your finger. A tap on empty ground closes the card.
The two arrows in the header, or ⌘Z and ⇧⌘Z. A slider drag is one step. On narrow screens they move into the ⋯ menu.
vol is the output level, from 0 to 100. It isn't saved with a preset. Panic, the round button beside it, clears every echo and the feedback at once (esc). The ⋯ menu repeats clear delay, play demo and stop source, and holds hide ui, this guide, the intro and what's new.
Choose hide ui in the menu, or press i. Everything but the valley fades and the sound carries on. Tap anywhere, or press i or esc, to bring the controls back.
Most of Klangtal is played on the water. Tap something and its card opens: top right on a wide screen, as a sheet at the bottom on a phone. esc or ✕ closes it.
When the tap's echo sounds, in ms and in beats at the tempo. The card's title shows the same, as 82 ms · 0.13 beats. Dragging the buoy does the same thing.
− and + move it 1 ms earlier or later, or one grid point when it snaps.How often this tap lets new sound in. On every step the tap rolls a die: at 50% it takes new sound about every other step. Echoes already in the lane are left alone.
A little chance on every tap keeps a pattern from repeating exactly.Switches the tap off without removing it. An off buoy is dimmed and keeps its place.
The same switch as tap snap in the delay tab. While the delay is synced it always snaps, and the switch can't be changed.
Takes the tap away. A lane keeps at least one. The other buoys keep their echoes.
Adds a buoy to the same lane and selects it. The count beside it reads 4 of 8.
With a buoy focused from the keyboard, the arrow keys move it 1 ms (10 ms with ⇧), delete removes it and return opens its card. When it snaps, the arrows move it one grid point (four with ⇧).
Each lane has its own copy of every effect it uses, as a label on the channel beside its river. Lane A's are on the left, lane B's on the right.
How much of the effect you hear. The same as the effect's slider in fx, and shared by both lanes, with its settings below it.
Bypasses the effect in this lane only. A bypassed label is struck through and dashed, and the effect keeps its place.
Moves the effect to another zone, at its end. A zone that is full is greyed out until you move or remove something there.
Moves it one place along its zone, so the sound passes through it sooner or later. The line beside them reads 1 of 1 before.
Takes the effect out of this lane only. The other lane keeps its copy.
The cross slider: how much of each lane's returning echoes goes to the other lane. The pill reads cross 26%.
The same switch as zigzag in the delay tab. The pill adds · zigzag, and dotted arcs cross the ridge.
Tap A or B beside a river to open rhythm with that lane marked. While link lanes is on, a small chain sits above the cross pill and lane B's buoys are dimmed: they follow lane A.
The valley's light and weather are part of the look, saved with your session and with every preset. Set them with weather and time of day in the delay tab. With weather following the dirt, clouds and rain come in as the sound gets dirtier and clear slowly when it cleans up.
How the two rivers of echoes behave: their length, how much comes round, how they cross, and the sky.
Locks the length of both lanes to the tempo. time gives way to delay beats and tempo, and the line under swirl shows the length, as 1277 ms at 94 bpm. Free, it reads free time.
Synced buoys settle onto 1/16 notes and triplets as you drag them.The length of the lanes while synced. Straight: 1/16, 1/8, 1/4, 1/2, 1 bar, 2 bars, 4 bars. Dotted: dotted 1/8, dotted 1/16, dotted 1/4, dotted 1/2. Triplet: 1/16 triplet to 1/2 triplet.
Shown here while the delay is synced. It is the same tempo as in rhythm.
The length of the lanes in free time. Together with stretch it sets how late the furthest buoy can be.
Scales the gaps between the buoys. Low, every echo crowds close behind the sound. High, they spread over the whole length.
How much of each echo comes round again. Low, a few echoes. High, a long tail that every feedback effect works on again on each pass.
A guard holds the level back if the echoes build up too far, so high settings stay playable.How much of each lane's returning echoes go to the other lane. At 0 the lanes stay apart. At 100 every return swaps sides.
Every other buoy listens to the other lane: A2 and A4 hear lane B, B2 and B4 hear lane A. Echoes then step back and forth across the ridge as they repeat. A buoy's card says Reads lane B when it does.
The balance of the source and the echoes. At 100 you hear only echoes.
Puts the buoys on a grid of sixteen points along the lane, so each lands on a sixteenth of the delay.
Pans lane A to the left and lane B to the right. At 0 both are in the middle.
Turns each returning echo round the stereo field, lane A one way and lane B the other. With feedback up, the tail circles.
Follows the dirt: a storm gathers as grit, drive, stutter and speaker breakup get louder in the sound, and clears when they stop. Or keep it calm or stormy. The sound doesn't change.
The light in the valley. It fades over about a second. Shown as day to start with.
The amount and settings of every effect, for both lanes. Where an effect sits, and whether a lane uses it, is set on the water (see effects). Each effect with settings has a settings line under its slider that opens them.
Locks tremolo, phase and stutter to the tempo. Each swaps its rate for a beat value of its own, and shows what that is, as 1.57 Hz at 94 bpm. It is separate from delay sync.
Plays the sound through a small speaker that goes from clean to cracking. Low, it narrows the sound. High, it breaks up, with the odd crack when pushed. New setups put it before the delay.
After the delay it breaks up the whole tail. Inside feedback it eats each echo a little more.The speaker: how much of the top it keeps and how soon it breaks up. radio to start with. Tannoy and radio keep the least of the top and cube amp the most. Blown 12 breaks up soonest and cube amp last.
How hard the sound hits the speaker. More drive breaks it up sooner, and the level is brought back down after it, so it gets dirtier, not louder.
A dirty filter in one slider. More grit pushes the sound harder and closes the top, at about the same level.
Inside feedback, each echo comes back rougher and darker than the one before.Pulses the level. In free time it speeds up a little when the sound is louder. At 100 the level nearly vanishes at the bottom of each pulse.
How fast it pulses, in free time.
One pulse in beats, while sync fx is on. Starts at 1/4.
A sweeping phase filter. Inside feedback the sweep is a little further on each time an echo comes round.
How fast it sweeps, in free time.
One sweep in beats, while sync fx is on. Starts at 1 bar.
Catches a short slice of the sound and repeats it three times, then catches the next. The slider is how much of the repeats you hear.
How long each slice is, in free time. It is separate from the amount.
The slice length in beats, while sync fx is on. Starts at 1/16.
A reverb tail of up to about seven seconds. Inside feedback every echo gains more room as it repeats.
Wow, flutter, warmth and a little hiss, all together, so the sound wavers and wears like an old tape. New setups leave it out: add it from an empty socket.
Inside feedback each pass wears the echo a little more. At high amounts the tail is shorter than a clean one.A slow chorus: three voices round the sound, a little apart in time and pitch. New setups leave it out; it goes after the delay when you add it.
How fast the voices move, as a multiple of their own slow rates.
Filters the sound towards a vowel, so it sounds more like a voice. Inside feedback each echo comes round more vowel-like. New setups leave it out.
At high amounts the tail is shorter, more so than with the other effects.Which vowel. It moves smoothly between them as you drag.
Patterns for each lane: which steps let new sound in. Tap A or B in the valley to jump here with that lane marked.
The speed of the steps, two to a beat, and of every synced time.
Lane B copies lane A, and from then on an edit to either lane goes to both: taps, steps, length and effects. Switching it off leaves lane B as it is.
How many steps the lane's pattern has. The button reads length 8. Each lane counts its own, so lanes of different lengths drift apart and meet again.
Try 15 against 14, as in the preset fifteen against fourteen.Fills the steps in one go. Straight opens them all, sparse one in four, clusters three in every eight, random about half, and clear closes them all.
Tap a step to open or close it. An open (lit) step lets new sound into the lane. A closed step keeps it out, while the echoes already there carry on. The playing step is outlined.
Adds a buoy to the lane. Each buoy's chance is in its card.
A preset is everything except vol and the source: every slider and switch, both lanes' taps, steps and effects, and the sky.
Banks. Tap the preset's name in the header. There are 27 factory presets in seven banks:
Each says in a line what it does. The arrows beside the name, and [ and ], step through them. history holds the states you have passed through this session, and yours the ones you save.
Search and stars. The search line on top finds a preset by name in every bank. The star beside a preset puts it in a starred bank at the top of the list, in this browser only; tap it again to take it off.
The sky. Every factory preset has its own weather and time of day, so loading one changes the light. Your own presets save the sky you had.
Saving. Name it under save this preset and press save. Your presets stay in this browser.
Sharing. copy code makes a short code for the current sound, to paste in a message. paste code opens a box to load one. export file saves your presets to a file, and import file brings a file back.
Every note comes back a dotted eighth later, in time with the tempo. Set spread to 60% to split the two lanes left and right.
Each echo comes back darker, rougher and more worn than the last, until it dissolves into the room. Move grit to after in its card to hear the difference.
The two lanes let sound in on different counts, so the echoes interlock in a pattern that only repeats every 63 steps. Give two buoys a chance of 60% to loosen it.
Echoes step back and forth across the ridge and circle round the stereo field as they repeat. Lower swirl for a plain ping-pong.
Each echo comes back more like a sung vowel. Add ensemble after the delay for a wider choir.
| Key | Does |
|---|---|
| esc | Closes the open list or card. With nothing open, clears every echo (panic). With the controls hidden, brings them back. |
| [ ] | Previous and next preset. |
| i | Hide or show the controls. |
| ⌘Z ⇧⌘Z | Undo and redo. |
| ← → ↑ ↓ | On a focused buoy: 1 ms earlier or later, or one grid point when it snaps. With ⇧, 10 ms or four points. |
| delete | On a focused buoy: remove it. |
| return | On a focused buoy: open its card. |
An iPad version is planned. For now, open Klangtal in Safari on the iPad: the valley, the cards and the tabs all work by touch. Use headphones with the microphone.
Tap begin or the start pill: the browser keeps sound off until you do. Then press play, or pick a source. Check that vol is up, that mix isn't at 0 with nothing to echo, and that the steps in rhythm aren't all closed.
Check that mix is up, that the buoys are switched on, and that some steps are lit in rhythm. A tap with chance at 0 never takes new sound in.
Press panic or esc to clear the echoes, then lower feedback. With the microphone, wear headphones: speakers feed the echoes back into the mic.
The browser has to allow it, and only on a secure page (https). If you said no once, allow it again in the browser's site settings, then choose microphone under source.
That zone is full, or the lane already has every effect. Move one to another zone or remove it, and a socket appears.
An effect's amount and settings are shared by both lanes. Its place, bypass and removal are per lane. To make lanes differ, put the effect in a different zone in each, or remove it from one.
Each preset carries its own weather and time of day. Set them again in delay after loading.
Something missing or wrong? Email support@tinkerhaus.xyz, or see support.
Klangtal is inspired by multi-head tape echoes and rhythmic multi-tap delays. Its speaker, grit, tremolo, stutter, tape, ensemble, formant, phase and reverb are shared with the other Tinkerhaus apps.