livepilot 1.9.23 → 1.10.0

package/livepilot/skills/livepilot-core/references/sample-manipulation.md

@@ -0,0 +1,724 @@
+# Sample Manipulation — Advanced Techniques from the Masters
+
+Comprehensive knowledge base for professional sample manipulation. Each technique includes context, philosophy, and step-by-step Ableton workflow.
+
+---
+
+## 1. Granular Sampling Techniques
+
+### 1.1 Amon Tobin — Destructive Granular Layering
+
+**Description:** Tobin samples from diverse vinyl sources (jazz, Bollywood, film soundtracks, 1970s electronic music) — sometimes 80+ samples per track — then processes each through filters and modulation, re-records, chops again, and layers the fragments until the source is unrecognizable. The key insight: each resample pass destroys the original identity while preserving its energy.
+
+**When to use:** When you want dense, cinematic textures that sound like nothing else. When you have a sample that is "too recognizable" and needs to become raw material.
+
+**Ableton workflow:**
+1. Load sample into an audio track
+2. Add Auto Filter (bandpass, moderate Q) and Saturator (Sinoid Fold, subtle drive)
+3. Record output to a new audio track (Resampling input)
+4. Chop the resampled audio into 4-8 fragments using Simpler in Slice mode
+5. Rearrange slices into a new sequence — ignore the original order
+6. Record that arrangement to audio again
+7. Repeat steps 2-6 with different effects each pass (Grain Delay, Erosion, Shifter)
+8. After 3-4 passes, the material is fully transformed — layer the best fragments from each generation
+
+### 1.2 Burial — Hand-Placed Granular Collage
+
+**Description:** Burial works in SoundForge (a destructive waveform editor), dragging tiny audio snippets by hand with no grid, no quantization, no undo. Every element — R&B vocal fragments, UK garage drum hits, vinyl crackle, rain recordings, phone recordings of friends singing — is placed by ear. The "crackle" people associate with Burial is actually tape crackle pitched down an octave, not vinyl. His drums come from recording impacts on surfaces around his home, then processing them to resemble real drum kits.
+
+**When to use:** When you want organic, un-gridded rhythms with emotional depth. When the music needs to feel like a faded memory rather than a produced track.
+
+**Ableton workflow:**
+1. Collect source material: vocal a cappellas, field recordings, found sounds, self-recorded impacts
+2. Set Ableton's grid to OFF (Cmd+4 / Ctrl+4) — work without quantization
+3. Load each sound as a separate audio clip in Arrangement view
+4. Manually drag clips to position by ear — let things drift off-grid
+5. For Burial-style crackle: take a vinyl noise sample, duplicate it, crossfade the copies to avoid pops, consolidate, then pitch down 12 semitones (one octave) using Transpose
+6. Layer the pitched crackle under everything at -18 to -24 dB
+7. For ghost vocals: chop individual words from a cappellas, add heavy reverb (Decay 4-8s, Diffusion high, 60-80% wet), and tuck them at -15 dB in the background
+8. Fill dead space with atmospheric textures (rain, tape hiss, room tone)
+
+### 1.3 Four Tet — Found-Sound Granular Composition
+
+**Description:** Four Tet keeps sound collection and composition as separate activities. He records sounds on his phone (Hang drums, street sounds, conversations) then processes them later. His album "Rounds" was composed using copy-paste techniques without a quantized grid. The approach is about capturing the energy of a moment, then sculpting it into music.
+
+**When to use:** When working with field recordings or non-musical sources. When you want warmth and organic texture in electronic music.
+
+**Ableton workflow:**
+1. Dedicate sessions purely to sound collection — record everything interesting on your phone
+2. Import recordings into Ableton as raw audio
+3. Use Simpler in Classic mode with Warp ON — set warp mode to Texture (Grain Size 30-60, Flux 50-80%)
+4. Play the loaded sample across the keyboard — different notes reveal different textures
+5. Record improvised performances with these "instruments" to audio
+6. Build compositions by copy-pasting the best fragments in Arrangement view
+7. Add Grain Delay (Pitch +-12, Spray 50-100%, Frequency 40-80 Hz) for additional granular texture
+8. Process with Reverb and Delay to create space around the found sounds
+
+### 1.4 Ableton Grain Delay as Granular Engine
+
+**Description:** Grain Delay breaks incoming audio into tiny grains and plays them back with independent control over pitch, timing scatter, and grain rate. It is the closest thing to a granular synth in Ableton's stock effects.
+
+**When to use:** When you want to smear a sound into a texture cloud. When you need pitch-shifted ambience from a dry source.
+
+**Ableton workflow:**
+1. Load Grain Delay on an audio track or return track
+2. Set Dry/Wet to 100% (or use on a return for parallel blend)
+3. **Pitch**: 0 for pure texture, +-12 for octave-shifted clouds, +-7 for fifth harmonics
+4. **Spray**: 0 = tight rhythmic grains, 100+ = scattered ambient texture
+5. **Frequency**: Low values (1-20 Hz) = slow, evolving clouds. High values (80-150 Hz) = buzzy, rhythmic grains
+6. **Feedback**: 30-60% for evolving tails, 80%+ for self-generating textures (use with care)
+7. Automate Spray and Frequency over time for evolving granular landscapes
+8. Follow with Reverb (long decay, high diffusion) to smooth the grains into pads
+
+---
+
+## 2. Chopping and Slicing Philosophies
+
+### 2.1 J Dilla — Micro-Chopping and the Dilla Feel
+
+**Description:** Dilla's revolution was twofold: (1) micro-chopping — slicing samples into tiny fragments and rearranging them to create new melodic/harmonic phrases, and (2) disabling MPC quantization so every hit was placed by feel. The result: a "drunk" or "wonky" timing where kicks land 30ms early, snares drag 20ms late, and different elements have different swing amounts. Dilla spoke through his sample choices — each chop was selected to reinforce the emotional theme of the track.
+
+**When to use:** When you want humanized, soulful beats. When rigid quantization kills the groove. When building hip-hop, neo-soul, or lo-fi.
+
+**Ableton workflow:**
+1. Find a soul/jazz/funk record with a melodic phrase you love
+2. Load into Simpler, switch to Slice mode, set Slice By to "Manual"
+3. Place slice markers at musically meaningful points — the start of each note, chord change, or melodic fragment (NOT just transients)
+4. Right-click Simpler > "Slice to Drum Rack" to get each chop on its own pad
+5. **Critical step — disable quantization**: In the piano roll, turn grid to OFF or set to very fine (1/64)
+6. Sequence the chops by hand — drag each MIDI note slightly off-grid
+7. Apply different swing amounts to different elements: kick at 55% swing, snare at 62%, hats at 48%
+8. For the Dilla velocity feel: vary velocity between 80-127 on every hit — nothing at full velocity, nothing perfectly consistent
+9. Use Simpler's pitch control to transpose individual chops for new melodies from the original harmonic material
+
+### 2.2 Madlib — Eclectic Blind Crate Digging
+
+**Description:** Madlib buys records without previewing them, often based only on the cover art, instrumentation credits, or year. He does not limit himself to any genre — his collection spans dub reggae, avant-garde jazz, Brazilian music, African music, psychedelia, film soundtracks, classical, rock, and industrial. His secret weapon: flipping multiple sections from the same record (intro, bridge, outro) rather than just one loop. His beats have a cinematic, collage-like quality because the sources are so diverse.
+
+**When to use:** When your sampling sources have become predictable. When you want to break out of genre conventions. When building beat tapes or experimental hip-hop.
+
+**Ableton workflow:**
+1. Select a record/song you would never normally sample — the further outside your comfort zone, the better
+2. Listen to the ENTIRE track, not just the intro — mark interesting moments with Ableton's Set Cue Points (see `toggle_cue_point`)
+3. Sample 3-5 different sections from the same source
+4. Load each section into its own Simpler in Classic mode
+5. Pitch each section independently — try +-3 to +-7 semitones to obscure the source
+6. Layer sections from different parts of the same record — the intro chords under the bridge melody
+7. Add lo-fi processing: Redux (Downsample to 12-16, Bit Depth 10-12), subtle Saturator, Auto Filter with slow LFO
+8. Keep the arrangement loose — Madlib's beats breathe because they are not over-arranged
+
+### 2.3 DJ Premier — Stab and Loop Chopping
+
+**Description:** Premier's signature is the single melodic "stab" — a chord or melodic fragment chopped from a record, then replayed rhythmically to create a new groove. Unlike Dilla's micro-chops, Premier's chops are larger (full chord hits) and more percussive. He layers these stabs with heavy filtering and scratched vocal hooks. The philosophy: find the ONE moment in a record that has the most energy, isolate it, and build the entire beat around that moment.
+
+**When to use:** When you want hard, punchy sample-based beats. When building boom-bap or gritty hip-hop. When a single chord or stab has enough energy to carry a track.
+
+**Ableton workflow:**
+1. Find a record with a strong chord hit, horn stab, or melodic accent
+2. Isolate a 200-500ms fragment — the stab — and load into Simpler (One-Shot mode)
+3. Set the amp envelope: Attack 0ms, Decay 200-500ms, Sustain 0%, Release 50ms (tight, punchy)
+4. Sequence the stab rhythmically — 16th-note patterns with velocity variation
+5. Add EQ Eight: high-pass at 100 Hz (keep it tight), slight boost at 2-4 kHz for presence
+6. Layer with a filtered version of the same stab: duplicate the track, add Auto Filter (LP, cutoff 800 Hz) for a "dusty" double
+7. For the Premier vinyl character: add subtle Erosion (Noise mode, Amount 5-10) and Redux (Downsample 1.5-3x)
+8. Add a scratched vocal hook: load a vocal phrase into Simpler, map to keys, and "perform" scratch patterns by rapid note triggering with pitch bend
+
+### 2.4 RZA / Wu-Tang — Cinematic Soul Sampling
+
+**Description:** RZA's approach combines soul and kung-fu movie dialogue samples with gritty, detuned production. The philosophy: samples are not just musical — they are narrative. A dialogue sample sets the scene, a soul loop carries the emotion, and the drums (often from cheap drum machines run through distortion) provide the raw energy.
+
+**When to use:** When building narrative-driven beats. When you want samples to tell a story, not just provide harmony.
+
+**Ableton workflow:**
+1. Find a dialogue sample (film, interview, spoken word) that sets a mood
+2. Place it at the beginning or as an interlude — let it breathe without musical backing
+3. Find a soul/R&B loop with emotional weight — load into audio track
+4. Pitch down 2-5 semitones for the signature dark, detuned RZA feel
+5. Add heavy Saturator (Hard Curve, Drive 15-20 dB) for grit
+6. Program drums in Drum Rack — keep patterns simple, hard-hitting
+7. Process the entire drum bus through Overdrive (Drive high, Dynamics low) for that "blown speaker" sound
+8. Use scene-based arrangement: Scene 1 = dialogue intro, Scene 2 = beat drops, Scene 3 = verse, etc.
+
+---
+
+## 3. Resampling Chains
+
+### 3.1 Serial Resampling (Destructive Iteration)
+
+**Description:** Record a processed sound to audio, then process the audio again, then record again. Each pass accumulates artifacts, harmonics, and textural complexity. After 3-5 passes, the original sound is completely transformed. This is how Amon Tobin, Aphex Twin, and many IDM producers create sounds that cannot be reverse-engineered.
+
+**When to use:** When you want sounds nobody else has. When a synth patch sounds too "clean" or "stock." When you need to create a unique sonic identity for a track.
+
+**Ableton workflow:**
+1. Start with ANY sound — a synth note, a drum hit, a vocal phrase, a field recording
+2. **Pass 1 — Distortion character**: Add Saturator (Sinoid Fold) + Erosion → Record to new audio track
+3. **Pass 2 — Spatial smearing**: Add Reverb (100% wet, 3-5s decay) + Delay (dotted 1/8, 50% feedback) → Record to new audio track
+4. **Pass 3 — Pitch warping**: Load result into Simpler, transpose -12 or +12 semitones, enable Warp (Texture mode, Grain 20, Flux 80) → Record to new audio track
+5. **Pass 4 — Frequency sculpting**: Add Auto Filter (bandpass, resonance 60%, LFO to cutoff) + EQ Eight (aggressive notches) → Record to new audio track
+6. **Pass 5 — Final granular**: Add Grain Delay (Spray 100, Pitch +-5, Frequency 30 Hz) → Record to new audio track
+7. Compare all 5 generations — often generation 2 or 3 has the best balance of character and musicality
+8. Use fragments from different generations in the same track for textural variety
+
+### 3.2 Parallel Resampling (Wet/Dry Blend)
+
+**Description:** Instead of serial destruction, split the signal and process each path differently, then recombine. This preserves the original's clarity while adding complexity. The result is more controlled than serial resampling.
+
+**When to use:** When you want texture and character but need to maintain the original's rhythmic or melodic integrity. Good for vocals, leads, and any element that needs to stay recognizable.
+
+**Ableton workflow:**
+1. Create an Audio Effect Rack on the source track
+2. Create 3-4 chains:
+   - **Chain A (Dry):** No effects — 100% clean signal
+   - **Chain B (Destroyed):** Saturator (heavy) > Reverb (100% wet) > Auto Filter
+   - **Chain C (Pitched):** Grain Delay (Pitch +12, Spray 50) > Chorus-Ensemble
+   - **Chain D (Sub):** EQ Eight (LP at 200 Hz) > Saturator (Analog Clip) for sub harmonics
+3. Balance chain volumes: Dry at 0 dB, others at -6 to -12 dB
+4. Map chain volumes to Macro knobs for performance control
+5. Record the combined output to audio when you find a good balance
+6. Chop the best moments from the recording
+
+### 3.3 Effects-Tail Harvesting
+
+**Description:** Feed a short, percussive sound into a long reverb or delay, then capture only the TAIL — discard the original transient. The tail becomes a pad, texture, or atmospheric element with the spectral character of the original.
+
+**When to use:** When you need ambient material that harmonically relates to your existing sounds. When creating transitions or background layers.
+
+**Ableton workflow:**
+1. Create a return track with Reverb (Decay 8-15s, Diffusion max, Dry/Wet 100%, Pre-Delay 50-100ms)
+2. Send a short percussive sound (chord stab, snare, vocal syllable) to this return at 100%
+3. Record the return track output to a new audio track
+4. Trim away the first 500ms-1s (the initial transient bleed)
+5. What remains is a pad/drone with the harmonic character of the original
+6. Time-stretch this pad to fit your arrangement using Warp (Texture mode)
+7. For variation: repeat with different reverb settings or different source sounds
+8. Layer multiple tails at low volumes for rich, harmonically coherent atmospheres
+
+---
+
+## 4. Vocal Sampling Techniques
+
+### 4.1 Formant Shifting for Character Transformation
+
+**Description:** Formant shifting changes the resonant frequencies of a voice (which define whether it sounds male, female, large, small) without changing the pitch. Shifting formants UP makes a voice sound smaller/younger/feminine. Shifting DOWN makes it sound larger/deeper/masculine. This is different from pitch shifting, which changes both pitch and formants together.
+
+**When to use:** When you want to disguise a vocal source. When creating alien/robotic/otherworldly vocal textures. When a vocal sample is in the right key but the wrong "character."
+
+**Ableton workflow:**
+1. Load a vocal sample into an audio track
+2. For stock Ableton: use Simpler with Warp ON — set warp mode to Complex Pro
+3. Complex Pro's "Formants" parameter controls formant preservation — at 100% formants are preserved during transposition, at 0% they shift with pitch
+4. To shift formants independently: transpose the sample UP in Simpler, then set Formants to 0% — the pitch goes up AND the formants shift up, creating a chipmunk effect
+5. For more control: use Audio Effect Rack with two chains — one pitched up with formants at 0%, one pitched down with formants at 100% — blend for hybrid character
+6. For the "Prismizer" effect (Bon Iver): see Section 4.4 below
+7. Automate formant/pitch parameters over time for morphing vocal textures
+
+### 4.2 Time-Stretch Artifacts as Texture
+
+**Description:** Most producers try to minimize time-stretching artifacts. This technique deliberately maximizes them. When you extreme-stretch a vocal (200-1000% length), the algorithm creates granular clouds, metallic resonances, and ghost harmonics that become the texture itself.
+
+**When to use:** When you want haunting, ethereal vocal pads. When you need to transform a short vocal fragment into a long atmospheric layer. When building ambient or experimental music.
+
+**Ableton workflow:**
+1. Take a short vocal phrase (1-4 bars)
+2. Select the audio clip and change its Warp Mode to Texture (Grain Size 10-30, Flux 70-100%)
+3. Drag the clip's end to stretch it to 4-16x its original length
+4. The artifacts from extreme stretching become the texture — metallic ghosts, granular clouds, pitch smearing
+5. Alternatively: set Warp Mode to Tones and stretch 4x — Tones mode creates a more tonal, singing quality in the artifacts
+6. For maximum destruction: set Warp Mode to Beats (Transients mode, Envelope 10-30) and stretch 4x — this creates rhythmic stuttering from the vocal
+7. Layer the stretched vocal under the original at -10 to -15 dB for depth
+8. Add Reverb (100% wet, 5-10s decay) and Chorus-Ensemble to smooth artifacts into pads
+
+### 4.3 Vocal Chopping for Rhythmic Patterns
+
+**Description:** Chop vocals into individual syllables, phonemes, or even sub-syllable fragments, then sequence them as rhythmic percussion. The human voice has natural transients (consonants) and sustains (vowels) that map perfectly to percussive and melodic roles.
+
+**When to use:** When building vocal-driven electronic music (future bass, UK garage, house). When you want rhythmic energy from non-drum sources. When creating signature vocal hooks.
+
+**Ableton workflow:**
+1. Load a vocal recording into Simpler, switch to Slice mode
+2. Set Slice By to "Transient" with sensitivity around 50-70%
+3. Right-click > "Slice to Drum Rack" — each syllable/phoneme gets its own pad
+4. In the Drum Rack, process individual pads:
+   - Vowel sounds ("ah", "oh"): add Reverb, use for sustained/melodic parts
+   - Consonant sounds ("t", "k", "s"): keep dry, use as percussion
+   - Breaths: keep as rhythmic fills at low volume
+5. Sequence the chops into a pattern: consonants on beats 1 and 3, vowels on 2 and 4
+6. Apply Ping Pong Delay (1/16, 30% feedback) to the vocal percussion track for stereo movement
+7. Use Simpler's pitch control per-slice to create melodies from the vocal fragments
+8. Group the Drum Rack and add Glue Compressor to bind the chops together
+
+### 4.4 Bon Iver / Prismizer — Layered Harmonic Vocal Processing
+
+**Description:** The "Prismizer" effect (developed by Bon Iver collaborator Chris Messina) routes a vocal through a vocoder that uses the same vocal as BOTH carrier and modulator, combined with a harmonizer that creates parallel harmony voices. The result: a choir-like, crystalline vocal texture where the singer harmonizes with transformed versions of themselves. Heard on Bon Iver's "715 - CREEKS" and Frank Ocean's "Blonde."
+
+**When to use:** When you want a single vocal to sound like a choir of cyborg angels. When creating emotional, textured vocal-led tracks. When a simple vocal needs to become a wall of sound.
+
+**Ableton workflow:**
+1. Record a simple vocal melody (doesn't need to be perfect — imperfection adds character)
+2. Duplicate the vocal track 3-4 times
+3. On each duplicate, add a different pitch shift:
+   - Track 1: Original (no shift)
+   - Track 2: +3 or +4 semitones (major/minor third up)
+   - Track 3: +7 semitones (perfect fifth up)
+   - Track 4: -12 semitones (octave down)
+4. On each pitched track: set Warp to Complex Pro with Formants at 80-100% to preserve vocal character
+5. Add Vocoder on each track: set Carrier to "Noise" or "Modulator" for different textures
+6. Add Chorus-Ensemble to each track with slightly different Rate settings
+7. Add Reverb (Hall, 3-5s decay, 40-60% wet) on a shared return track
+8. Pan the layers: original center, third L30%, fifth R30%, octave center
+9. Automate the layer volumes over time — bring in the harmonies gradually
+
+### 4.5 Vocal Texture Freezing
+
+**Description:** Capture a single moment of a vocal (a vowel, a breath, a consonant) and freeze it into an infinite sustain using reverb freeze, granular looping, or spectral hold. This single frozen moment becomes a drone or pad that carries the vocal's formant character indefinitely.
+
+**When to use:** When you need a pad that has human vocal quality. When transitioning between sections. When creating ambient beds from vocal material.
+
+**Ableton workflow:**
+1. Load a vocal into an audio track
+2. Add Reverb with Freeze enabled — set Decay very long (20s+), Diffusion max
+3. Play the vocal and engage Freeze (automate the Freeze parameter) at the exact moment you want to capture
+4. The frozen reverb sustains that vowel/consonant indefinitely
+5. Record the frozen output to a new audio track
+6. Use this recording as a pad — add Auto Filter with slow LFO for movement
+7. Alternative: load a vocal into Simpler, find a sustaining vowel, set very short loop points (10-50ms) with crossfade — this creates a granular freeze of that moment
+8. Transpose the frozen texture across the keyboard for chords
+
+---
+
+## 5. Texture Creation from Samples
+
+### 5.1 Paulstretch-Style Extreme Stretching
+
+**Description:** Paul's Extreme Sound Stretch uses spectral smoothing with phase randomization to stretch audio 10x-1000x its original length without artifacts. Unlike standard time-stretching, Paulstretch destroys all temporal structure (rhythm, transients) while preserving spectral character (timbre, harmony). Any sound becomes an ambient drone.
+
+**When to use:** When you need ambient pads or drones from any source material. When creating transitions or atmospheric layers. Works best on harmonically rich material (chords, melodies, voices) — drums and percussive sounds lose their character.
+
+**Ableton workflow (approximating Paulstretch with stock tools):**
+1. Take a harmonically rich sample (a chord, a vocal phrase, a melodic fragment) — 2-10 seconds long
+2. Load into an audio track, set Warp Mode to Texture
+3. Set Grain Size to maximum (the higher the smoother), Flux to 50-70%
+4. Stretch the clip to 10-20x its original length by dragging the clip end
+5. The result: a long, evolving drone that retains the harmonic character of the source
+6. For even smoother results: bounce this to audio, then stretch AGAIN (double-stretch)
+7. Add Reverb (100% wet, 8-15s decay) and Chorus-Ensemble to smooth granular seams
+8. For true Paulstretch quality: use the free PaulXStretch plugin as an external effect, or process offline in Audacity (which has Paulstretch built in) and reimport
+
+### 5.2 Reverse-and-Layer
+
+**Description:** Reverse a sample, process it, then layer it with the original forward version. The reversed version provides a "pre-echo" or "inhale" effect before each attack. When layered with the original, it creates a breathing, living quality where sounds seem to both arrive and depart simultaneously.
+
+**When to use:** When creating cinematic transitions. When a sound needs to feel like it exists in a time-warped space. Essential for reverse reverb effects.
+
+**Ableton workflow:**
+1. Place your sample in Arrangement view
+2. Duplicate it to a track below
+3. Select the duplicate clip > Reverse (press R or right-click > Reverse)
+4. Process the reversed version: add Reverb (100% wet, 4-8s decay), then consolidate/freeze
+5. Reverse the processed clip AGAIN — now the reverb tail precedes the original sound
+6. Align the reversed-reverb tail so it builds into the attack of the original
+7. Crossfade the two at the meeting point for a smooth transition
+8. For Simpler-based approach: load sample, press Reverse button, record the output with effects, reimport
+
+### 5.3 Creating Ambient Pads from Drums
+
+**Description:** Take a drum hit (kick, snare, cymbal) and use extreme processing to reveal and sustain its hidden harmonic content. Every acoustic drum sound contains pitched resonances — the trick is to expose them.
+
+**When to use:** When you want pads that are harmonically "native" to your drum sounds. When building tracks where every element derives from the same source material. When you need pads that blend naturally with your drum bus.
+
+**Ableton workflow:**
+1. Load a drum hit (snare or cymbal work best — they have the richest harmonics) into Simpler
+2. Set Warp Mode to Texture (Grain Size 80-120, Flux 40-60%)
+3. Stretch the hit to 10-20x length — the transient disappears, leaving the resonant body
+4. The result is a tonal drone at the drum's natural pitch
+5. Add Reverb (100% wet, 8-15s decay) to smooth the texture
+6. Add Auto Filter (LP, cutoff 2 kHz, slow LFO) to create gentle movement
+7. Play this across the keyboard — the drum's resonance transposes to create a tonal palette
+8. Layer multiple drum-derived pads (kick-pad for sub, snare-pad for mid, cymbal-pad for air) for a full-range texture bed
+
+### 5.4 Spectral Processing and Freezing
+
+**Description:** Use FFT-based spectral processing to isolate, manipulate, or freeze specific frequency components of a sound. This includes spectral blur (smearing frequencies over time), spectral freeze (holding a single spectral snapshot), and spectral filtering (surgically removing or boosting narrow frequency bands).
+
+**When to use:** When you need surgical control over a sound's frequency content. When creating evolving textures from static material. When doing advanced sound design for film/game audio.
+
+**Ableton workflow (using stock tools to approximate spectral techniques):**
+1. **Spectral freeze approximation**: Use Reverb's Freeze function on a sound — this captures and holds the current spectral content indefinitely
+2. **Spectral blur**: Add Corpus (resonator) after your sound — it imposes resonant frequencies that blur the original into a tonal cloud. Tune Corpus to the track's key.
+3. **Spectral filtering**: Use EQ Eight with very narrow Q values (Q > 10) to isolate specific harmonics of a sound. Boost one harmonic +15 dB, cut everything else. The result: a sine-like tone at that harmonic's frequency derived from the original.
+4. **Spectral layering**: Duplicate a track 3 times. On each, use EQ Eight to isolate a different frequency band (0-500 Hz, 500-2000 Hz, 2000+ Hz). Process each band independently, then recombine.
+5. For true spectral processing: use Max for Live spectral devices (Spectral Blur, Spectral Time) or external plugins like iZotope Iris, Paulstretch, or FluCoMa tools
+
+---
+
+## 6. Sample Layering for Composition
+
+### 6.1 Key and Tempo Detection for Musical Layering
+
+**Description:** Before layering samples, detect their key and tempo so they can be musically aligned. Modern key detection uses chromagram analysis — converting audio's frequency spectrum into 12 pitch classes and correlating them with major/minor key profiles (Krumhansl-Schmuckler algorithm). Tempo detection uses onset detection functions and autocorrelation to find periodic beat patterns.
+
+**When to use:** Every time you layer samples from different sources. Essential for avoiding dissonant clashes between samples.
+
+**Ableton workflow:**
+1. **Tempo detection**: Ableton auto-detects tempo when you drag audio in — verify by listening. For manual check: find two consecutive beats, measure the time between them, calculate BPM = 60 / (time in seconds). Use `read_audio_metadata` to check embedded tempo.
+2. **Key detection**: Use `get_detected_key` (LivePilot tool) which uses chromagram analysis. Or: listen and use `identify_scale` to verify by ear.
+3. **Matching**: Once keys are known, transpose the non-matching sample to fit:
+   - Same key = perfect match
+   - Related key (relative major/minor, dominant, subdominant) = usually works
+   - Distant key = transpose by the interval between the two keys
+4. Use Simpler's Transpose to adjust pitch without affecting timing (when Warp is ON)
+5. For tempo matching: Warp the sample and set the correct original BPM — Ableton handles the rest
+6. **Programmatic approach**: Chromagram = STFT > map to 12 pitch classes > sum energy per class > correlate with Krumhansl-Schmuckler profiles for all 24 keys > highest correlation = detected key
+
+### 6.2 Complementary Frequency Layering
+
+**Description:** Layer samples so each occupies a different frequency band, creating a full-spectrum composition from found sounds. Think of it as an orchestra where each sample is a different "instrument section."
+
+**When to use:** When building compositions from diverse sample sources. When you want each layer to contribute without masking others.
+
+**Ableton workflow:**
+1. Categorize your samples by frequency content:
+   - **Sub/Bass layer** (20-200 Hz): bass notes, low drones, pitched-down material
+   - **Mid layer** (200-2000 Hz): melodic content, chords, voices
+   - **Presence layer** (2000-6000 Hz): attacks, articulations, percussive elements
+   - **Air layer** (6000-20000 Hz): textures, shimmer, noise, ambience
+2. Load each sample on its own track
+3. On each track, add EQ Eight to carve its frequency space:
+   - Bass layer: LP at 200 Hz
+   - Mid layer: BP 200-2000 Hz
+   - Presence layer: BP 2000-6000 Hz
+   - Air layer: HP at 6000 Hz
+4. Match all samples to the same key and tempo (see 6.1)
+5. Build the arrangement by introducing layers one at a time
+6. Use `get_masking_report` to check for frequency conflicts between layers
+
+### 6.3 Creating Counterpoint from Found Sounds
+
+**Description:** Found sounds (field recordings, environmental audio, household objects) can be pitched and arranged into contrapuntal musical lines. The key insight: every sound has a fundamental pitch — even a door slam or a glass clink. By detecting these pitches and building melodies from them, you create music that sounds both musical and alien.
+
+**When to use:** When making musique concrete or experimental compositions. When you want a completely unique melodic palette. When standard synths and samples feel too conventional.
+
+**Ableton workflow:**
+1. Record 10-20 found sounds (tapping objects, environmental sounds, mechanical noises)
+2. Load each into Simpler in Classic mode with Warp ON
+3. For each sound, use `get_detected_key` or listen to find its natural pitch
+4. Set the Simpler's Root Note to match the detected pitch — now MIDI C3 plays the sound at its natural pitch, and other keys transpose it accurately
+5. Write a melody using each found sound as an "instrument"
+6. For counterpoint: assign different found sounds to different melodic lines
+   - Line 1 (glass taps): plays the main melody
+   - Line 2 (metal scrapes): plays a counter-melody in contrary motion
+   - Line 3 (wood knocks): plays a bass line
+7. Process each line with appropriate effects (reverb for cohesion, EQ for frequency separation)
+8. The result: a fully musical composition where every sound is sourced from the real world
+
+---
+
+## 7. Ableton-Specific Techniques
+
+### 7.1 Simpler vs Sampler — When to Use Which
+
+**Simpler** is the right choice when:
+- Working with a single sample
+- You need warping (Sampler cannot warp)
+- You want Slice mode (Simpler exclusive)
+- Quick sample playback with basic processing
+- Creative mangling with warp modes
+
+**Sampler** is the right choice when:
+- Building multi-sample instruments (mapping different samples across velocity and key zones)
+- You need advanced modulation matrix (multiple LFOs, envelopes mapped to any parameter)
+- Creating realistic acoustic instrument patches from multiple recordings
+- You need sample crossfading between velocity layers
+
+**Key difference**: Simpler has warping and slicing. Sampler has multi-sample zones and deeper modulation. For creative sampling, Simpler is almost always the right starting point.
+
+### 7.2 Simpler Modes Deep Dive
+
+**Classic Mode:**
+- Standard sample playback with ADSR envelope
+- Supports warping — sample plays in time regardless of pitch
+- Best for: melodic instruments, pads, one-shot effects, pitched playback of any sample
+- Tip: Enable Loop and set short loop points (10-50ms) on a sustaining portion for granular-style freeze
+
+**One-Shot Mode:**
+- Trigger-and-forget — sample plays through completely regardless of note length
+- No ADSR (just Fade In and Fade Out)
+- Best for: drum hits, percussion, sound effects, stabs
+- Tip: Use Fade Out to control tail length. Map velocity to volume for dynamic playing.
+
+**Slice Mode:**
+- Non-destructively divides sample into playable slices
+- Slice By options: Transient, Beat, Region, Manual
+- Best for: drum breaks, vocal phrases, melodic loops
+- Tip: After slicing, right-click > "Slice to Drum Rack" for individual pad control per slice
+
+### 7.3 Warp Modes as Creative Effects
+
+Each warp mode produces different artifacts when pushed beyond its intended use:
+
+**Beats Mode (for creative abuse):**
+- Designed for: rhythmic material with clear transients
+- Creative abuse: Apply to non-rhythmic material (pads, drones) and stretch 2-4x
+- Result: The algorithm imposes a rhythmic stutter on non-rhythmic material, creating glitch patterns
+- The "Preserve" dropdown (Transients/Off/1/4/1/8/etc.) determines the repeat grid
+- Envelope parameter controls how much of each "slice" plays — low values = choppy, high = smooth
+
+**Tones Mode (for creative abuse):**
+- Designed for: monophonic melodic material
+- Creative abuse: Apply to polyphonic or noisy material and stretch 2-4x
+- Result: Singing, warbling artifacts as the algorithm tries to find a pitch in complex material
+- Grain Size parameter: small values = metallic, robotic; large values = smooth, chorus-like
+
+**Texture Mode (for creative abuse):**
+- Designed for: ambient, textural, polyphonic material
+- Creative abuse: Apply to ANYTHING and use Grain Size and Flux as sound design tools
+- Grain Size: 1-20 = buzzy, granular, almost synthesis-like; 50-100 = smooth, cloud-like; 100+ = very smooth
+- Flux: 0 = stable, repeating; 50 = gently evolving; 100 = maximally random, chaotic
+- This is Ableton's closest stock equivalent to a granular synth
+- Tip: Load a 1-second drum hit, set Texture mode, Grain 10, Flux 80, stretch to 30 seconds = instant granular drone
+
+**Complex Pro Mode (for vocal work):**
+- Designed for: complex, polyphonic material
+- The Formants parameter (0-100%) controls how much formant character is preserved during pitch shifting
+- At 0%: formants shift with pitch (chipmunk/monster effect)
+- At 100%: formants stay natural while pitch changes (transparent transposition)
+- Envelope parameter controls transient preservation
+
+### 7.4 Slice to MIDI Workflow
+
+**Description:** Convert an audio loop into a MIDI-triggered Drum Rack where each slice is independently playable, rearrangeable, and processable.
+
+**Ableton workflow:**
+1. Load an audio loop (drum break, vocal phrase, melodic loop) into an audio track
+2. Double-click the clip to open it in the Detail View
+3. Adjust warp markers so the loop is properly in time
+4. Right-click the clip > "Slice to New MIDI Track"
+5. Choose slicing resolution:
+   - **Transient**: Slices at detected transients (best for drums)
+   - **Beat division** (1/4, 1/8, 1/16): Fixed grid slicing (best for rhythmic rearrangement)
+   - **Warp Markers**: Slices at your manual warp marker positions (most control)
+   - **Bar**: One slice per bar (best for long loops)
+6. A new MIDI track appears with a Drum Rack containing all slices
+7. The original MIDI pattern plays back identically to the audio — now rearrange, delete, and add notes
+8. Process individual slices: expand the Drum Rack, add effects per pad (reverb on snare slice, distortion on kick slice)
+9. Use `get_simpler_slices` to inspect slice points programmatically
+
+### 7.5 Simpler Creative Techniques
+
+**Reverse in Simpler:**
+- Toggle the Reverse button (or use `reverse_simpler`) to play the sample backwards
+- Combined with effects: Reverse + Reverb creates classic reverse-reverb swells
+- Use `warp_simpler` to enable/disable warping on the reversed sample
+
+**Crop and Replace:**
+- Use `crop_simpler` to trim a sample to only the visible region — useful after finding the perfect section
+- Use `replace_simpler_sample` to swap samples while keeping all Simpler settings (envelope, filter, etc.)
+
+**Sample Start Modulation:**
+- Map an LFO or envelope to Simpler's Sample Start parameter
+- Each note triggers from a slightly different position in the sample
+- Creates natural variation — no two notes sound identical
+
+---
+
+## 8. Key Detection Approaches
+
+### 8.1 Chromagram Method (Most Common)
+
+**Algorithm:**
+1. Compute Short-Time Fourier Transform (STFT) of the audio signal
+2. Map each frequency bin to one of 12 pitch classes (C, C#, D, ... B) by folding all octaves together
+3. Sum the energy in each pitch class across time to create a Pitch Class Distribution (PCD)
+4. Correlate the PCD against 24 key profiles (12 major + 12 minor) using the Krumhansl-Schmuckler profiles
+5. The key with the highest correlation coefficient is the detected key
+
+**Strengths:** Simple, fast, works well on tonal music with clear harmonic content.
+**Weaknesses:** Struggles with atonal music, music with many key changes, or heavily percussive material with weak pitch content.
+
+**LivePilot integration:** The `get_detected_key` tool performs chromagram-based key detection on any audio clip, and `get_chroma` returns the raw 12-dimensional chroma vector for further analysis.
+
+### 8.2 Hidden Markov Model Method
+
+**Algorithm:**
+1. Compute chromagram as above
+2. Define states as 24 keys (12 major + 12 minor)
+3. Define transition probabilities between keys (e.g., C major to G major = high probability, C major to F# major = low probability)
+4. Use the Viterbi algorithm to find the most likely sequence of keys over time
+5. The global key is the most frequent state
+
+**Strengths:** Handles key changes and modulations. More accurate for complex music.
+**Weaknesses:** More computationally expensive. Requires training data for transition probabilities.
+
+### 8.3 Practical Key Detection in Ableton
+
+**Quick ear-based method:**
+1. Find a sustained note or chord in the sample
+2. Use a tuner (Ableton's Tuner device on the track) to identify the pitch
+3. Play along with a piano/synth to find which scale fits
+4. Use `identify_scale` with the notes you hear to confirm
+
+**Hybrid method:**
+1. Run `get_detected_key` for algorithmic detection
+2. Verify by ear — play the detected key's scale over the sample
+3. If the algorithm is wrong (common with ambiguous material), try the relative major/minor or the parallel key
+4. Use `classify_progression` on any identified chord sequence for deeper harmonic analysis
+
+---
+
+## 9. Transient Detection and Musical Slicing
+
+### 9.1 Transient Detection (Energy-Based)
+
+**Algorithm:**
+1. Compute the spectral flux: for each STFT frame, calculate the sum of positive differences between consecutive magnitude spectra
+2. Apply a peak-picking function with adaptive threshold to find onset candidates
+3. Filter candidates by minimum inter-onset interval to avoid double-triggers
+4. Each detected onset marks a transient
+
+**Ableton integration:** Simpler's Slice mode with "Transient" option uses onset detection. The Sensitivity parameter controls the adaptive threshold — higher = more slices, lower = fewer.
+
+**LivePilot integration:** `get_onsets` returns detected onset positions for any audio clip.
+
+### 9.2 Beat-Aligned Slicing
+
+**Description:** Slice at metrically meaningful positions (beats, half-beats, bars) rather than at acoustic transients. This is better for rhythmic rearrangement because every slice aligns to a musical grid.
+
+**When to use:** When rearranging drum breaks. When you want to swap beats between different loops. When building remix-style arrangements.
+
+**Ableton workflow:**
+1. Load a loop into Simpler, switch to Slice mode
+2. Set Slice By to "Beat" and choose the division (1/4 for beats, 1/8 for half-beats, 1/16 for fine)
+3. Ensure the sample is correctly warped (BPM matches) so beat divisions align to actual beats
+4. Slice to Drum Rack for per-beat access
+5. Now you can rearrange beats: play beat 3 where beat 1 was, reverse beat 4, etc.
+
+### 9.3 Phrase-Level Segmentation
+
+**Description:** Detect higher-level musical boundaries — verse/chorus transitions, 4-bar phrase boundaries, section changes. This goes beyond beat detection into music structure analysis.
+
+**When to use:** When working with full songs or long recordings. When you need to extract specific sections (just the chorus, just the bridge). When planning arrangement structure.
+
+**Ableton workflow:**
+1. Load the full track into Arrangement view
+2. Use `get_arrangement_clips` to see the current arrangement structure
+3. Use Ableton's Set Cue Points at section boundaries (see `toggle_cue_point`)
+4. Use `analyze_composition` for AI-powered structure detection
+5. Use `get_section_graph` to see detected section boundaries and their relationships
+6. For manual phrase detection: look for energy changes, harmonic shifts, and arrangement changes in the waveform
+7. Place warp markers at phrase boundaries for easy slicing
+8. Use `get_phrase_grid` for algorithmically detected phrase structure
+
+### 9.4 Novelty-Based Segmentation
+
+**Description:** Detect points where the audio changes character — not just transients, but timbral shifts, harmonic changes, or textural transitions. Uses self-similarity matrices and novelty curves.
+
+**LivePilot integration:** `get_novelty` returns a novelty curve that highlights moments of significant change in the audio — peaks in this curve are musically meaningful slice points that go beyond simple transient detection.
+
+---
+
+## 10. Creative Constraint Techniques
+
+### 10.1 The One-Sample Challenge
+
+**Description:** Create an entire track using only one sample as the source for every element — drums, bass, melody, pads, effects. Forces extreme creativity in processing and resampling.
+
+**When to use:** When you are creatively stuck. When you want to develop deeper processing skills. When you want a track with absolute sonic cohesion.
+
+**Ableton workflow:**
+1. Choose a single sample (4-10 seconds of anything — a vocal phrase works well)
+2. **Kick**: Isolate a transient, pitch down 12-24 semitones, add Saturator for harmonics, shape with short ADSR
+3. **Snare**: Find a noisy/breathy section, pitch up 5-7 semitones, add Erosion (noise mode), short decay
+4. **Hi-hat**: Take the highest-frequency portion, pitch up 12+ semitones, add Redux (heavy downsample), very short decay
+5. **Bass**: Find a tonal section, pitch down 12 semitones, LP filter at 200 Hz, add Saturator (Analog Clip) for upper harmonics
+6. **Pad**: Stretch the full sample 10-20x using Texture warp mode, add Reverb (100% wet), Chorus
+7. **Lead**: Load into Simpler, set short loop points on a tonal section, play melodically across keyboard
+8. **FX/Transition**: Reverse the sample, add Grain Delay with high Spray, record and chop the best moments
+9. Use `apply_creative_constraint_set` to formalize constraint rules
+
+### 10.2 Genre-Locked Source Constraint
+
+**Description:** Sample only from a specific genre or era that is completely different from what you are making. Making techno? Sample only from country records. Making hip-hop? Sample only from classical recordings. The friction between source and destination creates unexpected results.
+
+**When to use:** When your music sounds too predictable. When you want to force yourself into unfamiliar harmonic and textural territory.
+
+**Ableton workflow:**
+1. Define your constraint: "I will only sample from [specific genre/decade/artist]"
+2. Select 3-5 recordings from that constrained source
+3. Listen through each, marking the most usable moments (unusual chords, interesting textures, surprising rhythms)
+4. Process heavily to obscure the source genre:
+   - Pitch shift +-5 to +-7 semitones
+   - Time-stretch 2-4x using Texture warp mode
+   - Heavy filtering (bandpass, sweep the center frequency)
+   - Resampling through distortion chains
+5. Build your target genre's structure using only these processed sources
+6. The constraint forces you to hear musical potential in unexpected places
+
+### 10.3 Time-Limited Composition
+
+**Description:** Set a strict time limit (30 minutes, 1 hour) to complete a track from sample selection to final arrangement. The time pressure eliminates overthinking and forces instinctive decisions.
+
+**When to use:** When perfectionism is killing your output. When you need to develop faster decision-making skills. When building a large body of work quickly (beat tapes, sketch albums).
+
+**Ableton workflow:**
+1. Set a timer (use your phone, not Ableton)
+2. First 5 minutes: select and load all samples — do not audition extensively, trust instinct
+3. Minutes 5-15: build the core loop — drums, bass, main harmonic element
+4. Minutes 15-25: arrange and add variation — verse/chorus structure, transitions
+5. Minutes 25-30: quick mix — volume balance, one EQ per track maximum, master limiter
+6. STOP when the timer ends — export as-is
+7. Resist the urge to "fix it later" — the rawness is the point
+
+### 10.4 Subtraction-Only Production
+
+**Description:** Start with everything and remove elements rather than building up. Load 8-16 loops/samples simultaneously, then sculpt the track by muting, filtering, and removing until only the essential elements remain.
+
+**When to use:** When your tracks feel thin or sparse. When you want to approach arrangement from a sculptural perspective. When working with large sample libraries.
+
+**Ableton workflow:**
+1. Load 8-16 loops/samples across tracks — all playing simultaneously
+2. Match all to the same key and tempo (see Section 6.1)
+3. Start with everything playing — it will be chaotic
+4. Begin removing: mute the least essential track. Listen. Mute another. Listen.
+5. Continue until every remaining element is essential — removing anything would be felt
+6. Now sculpt with EQ: on each remaining track, carve away frequencies occupied by other tracks
+7. Add subtle effects (reverb, delay) only where needed
+8. The arrangement is the process of reintroducing muted elements at different points
+9. Use `get_mix_issues` and `get_masking_report` to identify frequency conflicts for targeted removal
+
+### 10.5 Found Sound Only
+
+**Description:** No synths, no sample packs, no presets. Every sound in the track must be recorded by you from the physical world. Forces deep listening and creative processing.
+
+**When to use:** When you want a completely unique sonic palette. When developing your ear for the musicality in everyday sounds. When making musique concrete, experimental, or field recording-based music.
+
+**Ableton workflow:**
+1. Spend 1-2 hours recording sounds: kitchen utensils, doors, objects being struck/scraped/dropped, environmental ambience, your own voice
+2. Import all recordings into Ableton
+3. Audition each recording — note which have:
+   - Clear pitch (these become melodic elements)
+   - Strong transients (these become percussion)
+   - Sustained noise/texture (these become pads/atmosphere)
+   - Interesting rhythmic patterns (these become loops)
+4. Process each into its musical role using techniques from Sections 1-5
+5. Build a complete composition using only these processed field recordings
+6. The album "Plunderphonics" by John Oswald and the entire musique concrete tradition (Pierre Schaeffer) pioneered this approach
+
+---
+
+## Quick Reference: Technique-to-Tool Mapping
+
+| Technique | Primary Ableton Tool | Key LivePilot Tools |
+|-----------|---------------------|-------------------|
+| Granular texture | Grain Delay, Simpler (Texture warp) | `set_clip_warp_mode`, `warp_simpler` |
+| Sample chopping | Simpler (Slice mode), Drum Rack | `get_simpler_slices`, `crop_simpler` |
+| Resampling | Audio track (Resampling input) | `flatten_track`, `create_audio_track` |
+| Vocal processing | Complex Pro warp, Vocoder | `set_clip_warp_mode` |
+| Extreme stretch | Texture warp mode | `set_clip_warp_mode` |
+| Key detection | Tuner, `get_detected_key` | `get_detected_key`, `get_chroma`, `identify_scale` |
+| Transient detection | Simpler Slice (Transient) | `get_onsets`, `get_simpler_slices` |
+| Reverse | Simpler Reverse, Clip Reverse | `reverse_simpler` |
+| Formant control | Complex Pro warp (Formants) | `set_clip_warp_mode` |
+| Creative constraints | N/A (philosophy) | `apply_creative_constraint_set` |
+| Sample replacement | Simpler | `replace_simpler_sample`, `load_sample_to_simpler` |
+| Frequency analysis | EQ Eight, Spectrum | `get_master_spectrum`, `get_spectral_shape` |
+| Structure analysis | Arrangement view, Cue points | `analyze_composition`, `get_section_graph`, `get_phrase_grid` |