Ableton Sampler: The Complete Guide to Sampling and Sound Design

A short recording can become a playable bass, an evolving pad, a vocal lead, or an entire instrument. Ableton Sampler gives you control over which part of that recording plays, how it loops, how it responds to your keyboard, and how its tone changes over time.

The hard part is knowing which controls matter for the sound you want. A good loop can do more than another effect. A correctly assigned root note can save you from a patch that never plays in tune. And a small amount of modulation can turn a static sample into something you want to write music with.

This guide takes you from your first playable sample through multisampling and complete sound-design recipes. It expands John’s Ableton Sampler: The Complete Masterclass for EDMProd, with screenshots, animations, and listening excerpts from the session. Follow it from the beginning, or use the contents to jump to the sound you want to make.

Watch the full masterclass on YouTube. Audio excerpts come from the masterclass. Loud examples have been reduced in level; these are not loudness-matched A/B comparisons.

Before you begin: This guide covers Sampler, not Simpler or Drum Sampler. Sampler is included in Live 12 Suite; check Ableton’s edition comparison if it is missing from your browser. Use a MIDI track, a sample you can work with, and a comfortable monitoring level. If you are still learning Live’s tracks and browser, start with our Ableton beginner’s guide.

1. Sampler vs Simpler: choose the right instrument

Sampler turns recorded audio into a keyboard instrument with multisampling, flexible modulation, and extensive sound shaping. Its sample is the starting sound source. Envelopes, filters, and modulation then determine how that source behaves.

Simpler has its own strengths. In particular, its Warp and Slice features make it a useful choice for rhythmic material. Choosing between them is a workflow decision, rather than a progression from a beginner instrument to a professional one.

Understand the speed–pitch relationship

Play a sample an octave above its root and it plays twice as fast, taking half as long. Play it an octave below and it takes twice as long. In Sampler, those changes are linked.

That explains two things you will hear throughout this guide. A vocal becomes shorter as you play higher notes, and a chord made from a looping sample gradually drifts out of alignment because each note is playing the loop at a different speed.

You can use that drift creatively. If you need a phrase to retain the same timing at every pitch, choose a workflow with time-stretching instead.

Try this: Load a spoken word and play the same MIDI note in three octaves. Listen to the duration as well as the pitch. Understanding this behavior now will make tuning and looping much easier later.

2. Make your first playable instrument

Start with a short sound that has an interesting sustained section: a synth note, a vocal vowel, or a piece of a texture. A complete drum loop can work, but it makes your first tuning and looping decisions harder.

1. Load Sampler onto a MIDI track and drag your audio into its Sample display.
2. Arm the track and play a note. Confirm that you hear the sample before changing anything else.
3. Move Sample Start past any unwanted silence or slow introduction.
4. Enable a forward Sustain Mode loop, then position Loop Start and Loop End around a useful region.
5. Add a little Crossfade if the loop clicks or jumps.
6. In Filter/Global, shape the volume envelope. Use a short attack for a pluck, or a slower attack and release for a pad.
7. Tune the sample before writing a melody. The next section explains how.

John’s opening demonstration turns a sample into a sustained instrument, then reshapes it with filtering and modulation. The useful lesson is how quickly a small region can become the basis of a new patch.

Don’t add every effect immediately. First make a sound that starts cleanly, sustains as intended, and stops when you expect it to. Those three checks tell you whether the underlying instrument works.

Know where to find things

The Zone editor opens above the device. You can keep it visible while using the other tabs, which is useful when checking the tuning and playback settings of individual layers.

3. Tune the sample before designing the patch

A MIDI note name does not tell Sampler the pitch inside your recording. If your file contains a G but Sampler treats it as a C, playing C will produce G. Everything you write will be offset.

Root tells Sampler which MIDI note should play the recording at its original speed. It does not analyze or retune the recording for you.

A reliable tuning workflow

1. Bypass pitch modulation and any effects that make pitch hard to hear.
2. Put Live’s Tuner after Sampler.
3. Trigger the note currently shown in Root. At that key, you hear the source without keyboard transposition.
4. Identify the source’s pitch, including its octave. Set Root to that note.
5. Use the sample’s Detune for small tuning corrections.
6. Play several known MIDI notes and check that Tuner agrees.

If you know you recorded G1, set the sample’s root to G1. Don’t simply choose any G: the octave matters. John demonstrates this with a recorded sine wave, which makes the mismatch easy to see.

For an unfamiliar pitched recording, measure a stable portion rather than a noisy attack. If Tuner jumps around, temporarily loop the sustained part. A Spectrum display can provide supporting clues, but its biggest peak is not always the fundamental.

Root, Detune, Transpose, and Scale are different controls

Scale is not a musical-scale selector. At 100%, moving one semitone on the keyboard moves playback one semitone. At 0%, different keys can trigger the sample without changing its playback pitch—a useful option for certain percussion setups.

For chords and complex textures, choose a clear tonal reference and verify it musically. A sampled minor chord remains a transposed minor chord; it does not become a major chord because you play it over a major progression. Likewise, several peaks in Spectrum do not automatically prove that a sample contains several notes: one note already has harmonics.

4. Make loops that sustain cleanly

The Sample tab answers two separate questions: where does playback begin, and what happens when it reaches the loop boundary?

Sample Start and Sample End define the usable playback region. The sustain loop creates a repeating region inside it. You can retain a natural attack before the loop or start directly inside the sustained sound.

Forward or back-and-forth?

A forward loop reaches its end and jumps to its beginning. This is useful when the end and start join convincingly, particularly with periodic sounds.

A back-and-forth loop reverses direction at each boundary. It can create an appealing swelling motion in textures or short melodic fragments. It can also introduce an audible turn at the boundary, so listen before deciding that it is smoother.

Fix clicks in a sensible order

1. Move the boundaries away from obvious transients or abrupt changes.
2. Try Snap to place boundaries at zero crossings.
3. Add a small Crossfade and increase it only as needed.
4. Compare forward and back-and-forth playback.
5. Check several pitches, not just the note used while editing.

Snap is a starting point rather than a guarantee. The waveform can cross zero at both ends and still change direction abruptly. Stereo channels may also have different crossing points. Crossfade helps by blending the transition, but too much can alter the loop’s body or create a recurring swell.

For a sustained texture, move the whole loop to find a better region before trying to repair an unsuitable one. Sometimes a few hundred milliseconds farther into the recording makes the difference.

Link controls where the attack begins

Enable Link to make playback begin at Loop Start. Leave it off when you want the sample to play its original attack before reaching the sustained loop. This is especially useful when turning a texture into an immediate keyboard sound: the loop can start on the first note rather than after a long introduction.

Release behavior needs both sample playback and an envelope

The release settings in Sample determine where playback goes after note-off. The volume envelope’s Release determines how long you can hear it.

With Release Mode off, the sustain loop can continue while the volume fades. Other modes let playback continue toward Sample End, repeat a release region, or travel back and forth through it. Give the volume envelope enough release time to hear those differences.

The loop Detune controls can shift the sustain and release regions relative to the rest of the sample. Use them deliberately; an accidental setting can make a note change pitch when it reaches the loop.

Two display controls to remember: The waveform-height slider changes the display, not the audio level. The B/M/L/R buttons select the waveform view; displaying Mono does not convert the instrument’s output to mono.

5. Shape pitch and add FM or AM

The Pitch/Osc tab can turn a gentle sample into a punchy bass or a much more complex tone. Start with a tuned sine or another simple source so you can hear what each process adds.

Use a pitch envelope for movement within each note

Turn on P. Env, then set a nonzero Amount. With Amount at zero, changing the envelope shape will not bend the pitch.

A quick fall from a higher pitch into the played note adds a knock or kick-like attack. A slower rise produces a scoop. A long release movement can create a pitch tail, provided the volume envelope stays audible long enough.

Think of Amount as the range and the envelope’s levels as positions within that range. A useful first experiment is a positive amount, an immediate high peak, a short decay, and a sustain level that returns to the played pitch. Then adjust decay and curve while holding the rest steady.

For pitched instruments, always check where the envelope settles. A patch can have a tuned sample and still sustain sharp because its pitch envelope never returns to the intended note.

The extra oscillator modulates your sample

Sampler’s oscillator is a modulator, rather than another independently mixed sound source. In FM mode, it changes the sample’s frequency; in AM mode, it changes the sample’s amplitude. Raising its Volume increases the modulation depth.

Start with a sine modulation waveform and a low amount. Increase Volume gradually, then compare Coarse settings. Coarse is a frequency-ratio control, not a semitone transpose knob. Tune the source first if you want those relationships to behave predictably.

Fixed uses a fixed modulator frequency instead of following the played notes in the same way. Try it for metallic, bell-like, or deliberately inharmonic results, then test the patch across the range you will play.

The oscillator’s envelope lets you separate the attack from the sustained tone. A brief burst of modulation can emphasize the front of a bass. A sustained amount changes the whole note. A looped envelope repeatedly opens and closes the modulation, producing rhythmic movement.

Switch to AM and compare at a sensible output level. It is a different modulation process, not a guaranteed softer version of FM. The result depends on the source, waveform, frequency, and depth.

Envelope loops: Loop, Beat, Sync, and Trigger

A very short Loop Time does not necessarily create a very fast cycle if the attack and decay are still long. Change those stages together when designing a repeating contour.

Spread and glide

Spread creates detuned voices for width. It can flatter a pad or lead, but a bass deserves a mono check after adding it.

Glide gives monophonic sliding behavior. Portamento can work with polyphonic playback. The setting also affects how overlapping notes feel, so audition connected and separated MIDI notes rather than adjusting Time in isolation. Decide whether you want every note to have a fresh attack or a phrase that flows between pitches.

6. Control brightness, distortion, and articulation

The Filter/Global tab contains much of the instrument’s final character: its filter, waveshaper, volume envelope, panning, and voice count.

Start with the filter’s job

Use a low-pass filter to reduce brightness, a high-pass to remove low content, a band-pass to focus a region, or a notch to remove a band. The Morph filter blends between filter shapes and becomes especially interesting when modulated.

The 12 and 24 settings describe the filter slope in decibels per octave. Treat 24 as the steeper starting choice, rather than as a promise of exact attenuation at every frequency near the cutoff.

Resonance emphasizes the region around the filter’s characteristic frequency. Key tracking moves the filter with played pitch, helping a melodic patch retain a more consistent character across the keyboard. Velocity modulation can make harder notes brighter.

Turn on F. Env and raise its Amount when you want each note to open or close the filter. A short decay makes a pluck; a slow attack can reveal brightness gradually. Lower the base cutoff enough to leave room for the envelope to do something audible.

Choose a circuit by listening

Sampler offers Clean and several modeled circuits. Compare them using the same phrase, then compensate for output-level differences. A louder setting can seem better even when it is obscuring the attack or weakening the low end.

You do not need a hardware-emulation story to choose a useful setting. Listen for the tone, resonance, and response to Drive that suit the patch.

Change the shaper order

The shaper offers Soft, Hard, Sine, and 4bit characters. Try Soft for a first layer of saturation, Hard for a stronger edge, Sine for more dramatic folding, and 4bit for a coarse digital texture.

The routing arrow changes the order:

• Shaper → Filter: generate extra harmonic content, then shape it with the filter.
• Filter → Shaper: filter the source first, then let distortion create new content afterward.

Neither order is universally better. If the post-filter shaper brings back too much brightness, switch the order or add a final EQ after the instrument. Check the result with the shaper bypassed at a comparable listening level.

Shape the volume before adding space

For a pluck, lower sustain and set decay to the length you want. For a pad, allow sustain and use slower attack and release. A long release does not create audio past the end of an unlooped sample: the source must still be playing.

Set Voices for the musical role. A monophonic bass usually needs fewer voices than a sustained chord patch. Retrigger can keep repeated notes from accumulating as additional overlapping voices; with it off, long tails can build up in level.

The global Time control scales envelope timing. It is useful for changing a whole patch’s response, but it does not time-stretch the underlying audio file.

7. Add movement that serves the sound

Sampler has an auxiliary envelope and three LFOs. Use them to create one clear movement first, then add supporting variation.

Use Hz for a free-running rate or a beat division when the movement should follow the track. Retrig restarts the LFO for a new note. Offset changes the point in its cycle where it starts. Attack fades in the modulation over time—useful for vibrato that appears after a note has settled.

LFOs 2 and 3 also provide stereo options. Phase offsets the left and right modulation, while Spin lets their relationship evolve. Try these on a pad’s filter before applying them to the lowest part of a bass.

Change the starting point with Sample Offset

Dragging Sample Start is useful for editing, but it is not the same as having a directly automatable start control. John demonstrates the Sample Offset modulation destination as a way to vary where new notes begin.

1. Choose a sample with interesting differences along its length.
2. Set a sensible base Sample Start and enough Sample End range.
3. Assign Sample Offset to the auxiliary envelope or an assignable LFO.
4. Begin with a limited modulation amount.
5. Play repeated short notes and listen for different parts of the recording.

With the auxiliary-envelope approach, John controls the offset through the envelope’s Peak value, then automates that parameter. The internal LFO approach can introduce random variation without the external LFO device.

Sample Offset is evaluated for a new trigger. Moving it during a held note does not continuously scrub that note through the file like a granular scanning control. Loop Start and Loop Length are separate modulation destinations worth exploring for other kinds of movement.

For percussion, a small random offset can reduce repetition. For a vocal, a wide range might skip the useful vowel or land in silence. Set the range around the material you actually want to hear.

Make the patch playable

Use the MIDI tab to connect expression to sound. Try velocity to filter frequency, mod wheel to modulation depth, or pressure to a restrained amount of drive. Set the pitch-bend range intentionally so recorded bend data does what you expect.

A useful patch should respond well to playing, not just sound interesting while you drag controls. Test soft notes, hard notes, short notes, and a sustained phrase before adding more assignments.

8. Build a multisampled instrument

A single piano note stretched over the keyboard cannot reproduce all the changes in a real piano. Low notes have their own attacks and decay patterns; hard strikes have a different tone from soft strikes. Multisampling preserves more of those differences by using several recordings.

John builds a small example from C3 through C4, then adds a second velocity layer. It is a manageable way to learn the process before recording a much larger instrument.

Prepare the recordings

Record each required note at a consistent playing strength. Leave enough space for the sound to decay without overlapping the next recording. Trim unwanted silence and add a gentle ending fade if needed.

Name files so you can identify both pitch and layer—for example, C3_soft and C3_hard. Keep the names meaningful in the actual audio files, not just in temporary clip labels.

Repeat for another playing strength if the source changes character with velocity. A quiet sample is useful because it captures a different performance, not merely because its waveform is smaller.

Level matters: Preserve a sensible relationship between layers. Independently maximizing every recording can erase the dynamics you were trying to capture. Balance deliberately and test the result by playing.

Key zones choose the sample; Root determines its pitch

Load the recordings into Zone, then choose Key. Each horizontal range determines which incoming notes can trigger that sample.

For the octave exercise:

1. Set the C3 recording’s key range to C3.
2. Set its Root to C3.
3. Do the same for C♯3, D3, and the remaining recordings.
4. Play a chromatic scale and check every note with Tuner.
5. Extend the lowest and highest ranges only as far as the transposed results remain useful.

It is easy to complete the ranges but forget the roots. Then the correct file is selected and transposed incorrectly. If each successive key seems increasingly sharp, check Root before re-recording anything.

For recordings spaced farther apart, you can use Distribute Ranges Around Root Key after assigning correct roots. Still check the joins manually. Extending a range fills missing notes; it does not create newly recorded realism.

Velocity zones choose soft or hard performances

Switch to Vel. For a simple two-layer split, put the soft recordings at 1–64 and the hard recordings at 65–127. Test just below and just above the boundary.

If the ranges overlap without fades, both layers may play together in the overlap. If there is a gap, some velocities may produce nothing. Neither is inherently wrong when intentional, but both are common mapping mistakes.

John adds an on-screen correction to this part of the video: the overlapping piano layers sounded hollow because of their phase relationship. Treat that as a reason to check a blend by ear, rather than a target to copy.

To soften an obvious transition, overlap a small region and use the fade handles at the top of each zone. Listen to the blend: overlapping recordings can change level, phase, and tone. Constant-power fading is an option to compare, not an automatic guarantee of a transparent join.

Enable Auto Select when you want played notes to show you the active layers. This makes it easier to find the recording responsible for an unexpected attack or level jump.

Build only the detail the instrument needs

A first instrument can use a few well-chosen notes and two velocity layers. Add recordings where stretching sounds unnatural, where a tonal transition is too abrupt, or where the playing response feels flat.

The sample selector’s 0–127 range is not a 128-sample limit on the entire instrument. Avoid using that number to plan a library’s total capacity. For a broader reference, see Ableton’s Sampler manual and its multisampling walkthrough.

9. Build a sample chooser—or use round robin

Sample Select and round robin solve different problems. A selector lets you choose the sound; round robin introduces variation as notes arrive.

A macro for choosing kicks

Load several kick samples, open Sel, select their zones, and use Distribute Ranges Equally. Each kick occupies its own portion of the selector range.

Group Sampler into an Instrument Rack and map Sample Selector to a Macro. Now play one MIDI pattern and turn the Macro to audition different kicks in context.

For just two choices, you can assign selector positions 0 and 1 and limit the Macro’s mapped range accordingly. For five choices assigned to consecutive single positions, use 0–4. Name the Macro after its job, such as “Kick choice.”

The selector determines which samples are available when a note is triggered. It does not replace an already sounding sample halfway through its playback.

You can also store Macro Variations for useful choices. Remember that layers inside one Sampler share its broader sound-shaping setup. If you need entirely different filters, envelopes, and effect chains per sound, use separate rack chains.

Round robin for repeated hits

Record several claps that are similar but not identical. Trim each one, give them the same intended key range, and enable RR. Repeated MIDI notes can now cycle through the recordings instead of repeating one exact waveform.

Forward and Backward give an ordered cycle. Other chooses randomly without repeating the immediately previous sample. Random may repeat it. Use Reset when you want the cycle to restart at a musical interval.

Keep variations balanced enough that the rhythm remains steady. A little difference feels alive; one much louder or later clap can feel like a mistake. Add random pan only after the samples work together.

10. Recipe: turn a tiny loop into a synth

A very short repeating region stops sounding like a recognizable recording and starts behaving like an oscillator. This is the basis of John’s first sound-design example.

1. Load a sustained texture or drone.
2. Enable a forward sustain loop and turn on Link.
3. Shorten the loop until it contains only a small number of waveform cycles.
4. Move the loop to find a useful tone.
5. Tune the resulting loop with Root and Detune.
6. Shape its volume and filter envelopes as you would a synth patch.

The loop length now contributes directly to the pitch. Shortening it raises the repetition rate. Retune after a substantial loop edit, even if the original sample was correctly labeled.

A boundary discontinuity can become part of the tone at these speeds. You are no longer just fixing a slow, isolated click: you may be changing the harmonic content of the oscillator-like sound. Compare by ear instead of insisting on a perfectly smooth-looking waveform.

John adds pitch movement, filter shaping, Spread, and panning, then delay, reverb, and an arpeggiator. Keep the first version simple enough that you can hear the original tiny loop underneath those changes.

Your turn: Save three copies using different loop positions from the same file. Tune each, then play the same phrase. This gives you three related instruments rather than three random presets.

11. Recipe: make an evolving pad

For a pad, use a longer loop that retains some of the recording’s internal movement. John starts with a short melodic fragment, then layers an atmospheric texture.

Build the core layer

1. Find a relatively stable note or texture inside the sample.
2. Use a back-and-forth loop and Crossfade to smooth its movement.
3. Tune the usable region.
4. Set a slower volume Attack and Release, with enough sustain for a held chord.
5. Choose a Morph filter and assign a slow LFO to Filter Morph.
6. Add a restrained amount of FM if the tone needs more complexity.

Increase Spread and try stereo LFO movement after the center of the sound works. Check in mono before committing to a wide version. Use external filtering to leave room for the bass and avoid a pad that fills every register.

Add a contrasting layer

Open Zone and add a second sample. Give it a compatible root note, its own useful loop, and an appropriate sample volume. Solo or mute layers while checking their tuning and balance.

The second layer might contribute crackle, air, or a higher texture. It does not need to be as loud as the main layer. If it contains a chord, remember that its intervals move together when transposed; avoid accidentally building unwanted harmonies into every note you play.

Samples within one Sampler share its main envelopes and processing settings. Use separate Samplers in an Instrument Rack when each layer needs an independent filter or modulation design.

Finish with reverb after the dry sound works. Filter the reverb’s low end and compare a shorter decay before adding a huge tail. Long release plus long reverb can make every chord overlap the next one.

12. Recipe: build an 808 with a defined attack

Start with a tuned low sine or a sine-like sample. A slightly imperfect recording can add character, but do not mistake unwanted noise for a requirement.

An 808-style patch needs two parts to work together: a clear initial hit and a controlled low-frequency tail.

1. Check the sample’s root and fine tuning.
2. Remove an accidental starting click with the start point, Snap, or a small volume attack.
3. Set a decaying volume envelope with low sustain. Begin with a tail long enough to hear, then shorten it for the track.
4. Add a positive pitch envelope that falls quickly into the sustained note.
5. Try a brief burst of FM for more attack definition.
6. Use filter Drive or the shaper for harmonics, compensating with output level.
7. Set an appropriate voice count and test glide with your actual MIDI pattern.

The pitch envelope and FM envelope do different jobs. If the attack sounds like a laser, shorten or reduce the pitch movement. If it is too metallic, reduce FM depth or its duration. If a separate kick already supplies the transient, the bass may need much less attack shaping.

Add harmonics without losing the foundation

Drive can help the bass speak above the sub range. Compare Soft and Hard shaping, then try the alternate shaper/filter order. Stop periodically and listen with processing bypassed.

Avoid treating one harmonic pattern as a universal recipe for bass. Distortion behavior depends on the source, symmetry, amount, and surrounding processing. Choose the result that fits the kick and musical phrase.

John experiments with width and mid/side EQ. For a straightforward first patch, keep the deep low end centered and add width only where it contributes something useful. Check both stereo and mono after distortion; processing can change the balance you established earlier.

If you want more help placing the result in a track, read our sub-bass guide.

13. Recipe: build a moving growl with a separate sub

You do not need a complex starting sample to make an aggressive bass. John develops the growl from the same relatively simple source used for the 808, adding movement through modulation and distortion.

Begin with a short musical phrase so each change has context. Try the auxiliary envelope on Shaper Amount, then experiment with filter Drive and the modulation oscillator’s envelope. Small pitch-envelope changes can produce very different results once they pass through distortion.

Looping the modulation oscillator’s envelope can make repeated pulses that resemble a delay, even before a delay effect is added. Change attack, decay, and loop timing together to control that contour.

For a more aggressive finish, John adds multiband processing and saturation. These can expose upper detail and noise, so introduce them gradually and keep comparing output levels. Revisit earlier settings after adding them: the same FM depth may now be too intense.

Give the sub its own path

When the top becomes heavily distorted, moving, or wide, separate it from the foundation:

1. Group the sound into an Instrument Rack and create Sub and Top chains.
2. Keep the detailed processing on Top.
3. On Sub, remove the aggressive shaper, FM/AM, and unnecessary modulation.
4. Keep the musical pitch and amplitude behavior coordinated between the two.
5. High-pass Top after its distortion and other processing to remove unwanted low content.
6. Balance the chains while listening with the kick.

John uses a high-pass around 100 Hz in the demonstration. Treat that as a starting point for this patch, not a fixed crossover for every bass. Adjust for the notes, filter slope, phase relationship, and the remaining energy in both chains.

Adding another sample inside the same Sampler would not give you this independent processing. The rack’s separate instrument chains are what make the distinction useful.

14. Recipe: turn a vocal into a playable lead

Choose a sustained vowel or a small region with a clear pitch. You are building an instrument from part of the performance, so you do not need to preserve the timing of the whole phrase.

1. Load the vocal and identify the pitch of the chosen region.
2. Set Root and fine tuning.
3. Enable a sustain loop and move its boundaries around the useful vowel.
4. Add Crossfade to smooth the join.
5. Soften the volume attack slightly and add enough release for the phrase.
6. Try filtering, modest drive, and glide before adding spatial effects.

A very short loop can become buzzy; a longer loop preserves more of the singer’s movement. Both can be useful. If you move the loop to a different sung note, tune it again.

Use formant-style filtering for another layer of expression

John adds Live’s Auto Filter with its vowel-style filter options, then experiments with modulation. This is separate from Sampler’s internal filter. Use the available filter modes in your version of Live, and keep the movement narrow enough that the lead remains recognizable.

For controlled stereo delay, place a fully wet delay on one Audio Effect Rack chain and leave another chain dry. Put Utility after the delay if you want to reduce the width of the repeats without narrowing the dry lead. Balance the effect with the chain levels.

Keep reverb and feedback in proportion to the part. A sound that is impressive in isolation can become a wash once you play a melody. Record a phrase, then adjust the effects while it loops.

15. Recipe: make kicks, snares, and shakers from unexpected audio

The final section of the masterclass turns a synth effect into percussion. This is a useful exercise because it forces you to hear the ingredients of a drum: attack, body, noise, and decay.

Chapter note: The video’s final percussion demonstrations follow the vocal section. Use the timestamp links below to jump directly to the relevant examples.

Make a kick-like hit

Choose a source with enough body to survive heavy pitching. Keep it as a one-shot initially. Pitch it down, add a fast downward pitch envelope, and use a low-pass filter with a short envelope to shape the impact.

Add drive or shaping only after the transient and body are working. A final EQ can remove an unwanted rumble or a boxy region. If the source will not make a useful kick, try another sample or use the tiny-loop technique to build a more predictable oscillator-like source.

Make a snare-like hit

A snare needs enough body to register as an impact and enough noise to give it a tail. John reworks the same source, changes the filtering, and uses a noise waveform in the modulation oscillator to create additional texture.

Balance the initial click against the noisy part. A slightly delayed rise in the modulation can leave the transient clearer before the noise develops. If the sample remains too tonal, change its pitch, shorten the useful region, or choose a more suitable source.

Make a shaker and add variation

For a shaker, focus higher in the spectrum, soften the onset, and keep a short, useful decay. John retains noise modulation, filters out more lows, and experiments with shaping.

Then humanize carefully:

• Add small random changes to filter frequency.
• Vary Sample Offset for slightly different source material on each trigger.
• Add a restrained amount of random pan.
• Use different MIDI velocities so the pattern has accents.

A looping volume envelope can produce a repeating shaker from one held note. However, that is not the same as retriggering the sample with new MIDI notes: note-on-dependent changes such as Sample Offset will not refresh on every envelope cycle. Use a MIDI pattern when you want each hit to make those decisions again.

For more ideas about choosing and transforming source material, see our guide to sampling and free sample-pack collection.

16. Save a useful instrument—and finish something with it

Save a version as soon as the basic instrument works. Then save a second version after effects and performance controls are in place. It gives you a clean starting point if the patch becomes too complicated later.

Name the preset after what it does: “Vocal vowel glide,” “Soft two-layer piano,” or “Crackle pad.” If the sound depends on effects or multiple chains, save the containing rack as well as any individual Sampler preset you want to reuse.

Samples are external files. Keep them in a stable location, use Live’s file-management tools to collect the dependencies you need, and reopen the saved project before assuming it is portable. The definitive reference is Ableton’s Managing Files and Sets.

A short finishing exercise

Choose one sound from this guide and write eight bars with it. Add only the supporting parts needed to tell whether it works: a kick and rhythm for the bass, or a bass note and simple beat for the pad.

Automate one meaningful control. Print a short audio version. Then listen without touching the patch. Make the next adjustment because of what the music needs, rather than because another unused parameter is available.

17. Troubleshooting and quick answers

Can Sampler warp audio?

Sampler’s normal pitch changes alter playback speed and duration. Use Simpler with Warp or a suitable clip-based workflow when preserving timing independently of pitch is essential.

Can I turn any sample into a pitched instrument?

You can experiment with almost any recording, but not every source will give you a stable or useful pitch. A tiny loop can generate a tone; a longer loop can preserve texture. Tune the final region and test the musical range instead of relying on the original filename.

Why is my multisample still out of tune after mapping the keys?

The key zone chooses which recording plays. The sample’s Root determines how that recording transposes. Both need to be correct. Fine-tune individual recordings only after their roots are assigned properly.

How do I change the tone without changing the source sample?

Start with the filter and its envelope. Then compare the shaper and FM/AM oscillator. Add one clear modulation route at a time so you can tell which part of the patch is creating the change.

What should I learn next?

Make one instrument you want to keep, then use it in a track. Our Ableton workflow guide helps connect sound design to arrangement. For a different way to build your own sound sources, explore the Serum 2 wavetable-editor guide.

Based on John’s EDMProd masterclass, with supplementary checks against Ableton’s documentation. Screenshots, looping animations, and audio excerpts are from the source demonstration. Audio clips retain the session’s changing settings, with loud excerpts reduced in level for playback. They are listening examples, not controlled, loudness-matched comparisons.