Kinematic 0.5.0 is out. Modeling and sculpting in the viewport, a timeline on every graph, photo and sound editing. What's new
DevelopersDemosLearnCommunityVolunteerAboutDownload
Kinematic / Who it's for / Sound designers

Design your sounds in Kinematic.

Clean a field recording, turn a creak into a whoosh, then stretch, pitch and layer it right on the waveform, with every edit open to change. Lay it under the picture, set it in a room or fire it in a show. Free, even for commercial use.

  • Sound designers
  • Foley artists
  • Game audio
  • Post sound
  • Theater sound
  • Installation artists
Download for macOS 0.5.0See the work
Edits right on the waveformEvery edit stays openHosts CLAP and VST3macOS, Windows and Linux
On this page
Sound pane a creak becomes a whoosh
Stack every edit

A one second creak in eight steps, from the raw sound to a swell into the hit. Each step stays open in the Stack and sits on the canvas as a node.

The work

The sounds you make each week.

Cleanup, effects, voices, beds, synthesis and the delivery, and how each one goes in Kinematic. Every picture here is the app.

drips over a hum, before and after
01Clean

Clean a field recording.

Draw a range over a stretch of plain noise and choose Denoise: the hiss, hum or room tone under the whole take comes down and the sounds you wanted stay put. Take out rumble with a high pass, trim the bumps at the start and fade the end.

Clean›Denoise→Tone›High pass→Edit›Trim›Trim to range→Level›Gain and fades›Fade out
  • Remove hum at 50 or 60 Hz, a DC blocker and a noise gate for the gaps between footsteps
  • Compare any step with the one before it, at the same moment in the take
  • Record from any input, keep the take you like, and trim it in place
AudioDenoiseAudioHighPassAudioNoiseGateAudioClipAudioFade
Tutorial: clean up a field recording →Clean a field recording →
the Space menu open over a finished swell: reverbs, plates, delays, a convolver, flangers, phasers, a room simulator and a diffusor
the Space menu over a finished swell
02Design

Design whooshes, risers and swells.

Play a sound backward, stretch it to twice the length and give it a plate with a long tail. Turn it around again and the tail swells into the hit. The Space menu holds reverbs, plates, a spring, delays, flangers and a convolver that puts the space of any take on another sound.

Edit›Split, reverse and pass›Reverse→Time›Time stretch→Space›Plate with tail→Level›Loudness›Normalize peaks
  • Pitch it down for weight, or shift every frequency up for something stranger
  • Layer the plain sound back under it with Level > Layer > Mix with a take
  • Save the steps as a recipe and use them on the next sound
AudioReverseAudioTimeStretchAudioPlateReverbAudioConvolutionAudioPitchShift
Tutorial: turn a creak into a whoosh →Make a whoosh from a creak →
the Time menu open: time stretch, pitch shift, change rate, mono downmix, channel converter and frequency shifter
the Time menu: stretch, pitch and shift
03Voices

Shape voices and creatures.

Put a voice inside an old radio or a phone, drop it an octave for a monster, or let a voice shape a synth. Each filter, saturator and cabinet draws its curve in the viewport while you set it.

Tone›Band pass→Tone›Saturator→Tone›Cabinet simulator→Dynamics›Compressor
  • Pitch shift, a frequency shifter, a formant filter, a ring modulator and a vocoder
  • A bitcrusher for a cheap phone line, a room simulator for the space around it
  • Line a new take up to the guide track with the Dynamic warp node
AudioBandPassAudioCabinetSimAudioFormantFilterAudioVocoderAudioDynamicWarp
Tutorial: an old radio voice →Turn your voice into a synth →
the spectrum of one moment of a song, frozen into a drone and filtered
one moment frozen into a drone
04Textures

Build textures, drones and ambiences.

Scatter a recording into a cloud of grains, freeze one moment into a drone, or blend two sounds into one with cross synthesis. Cut eight even seconds of room tone with short fades, ready to loop under a scene.

Edit›Trim›Trim to range→Level›Gain and fades›Fade in→Level›Gain and fades›Fade out
  • Granular, a grain delay and a spectral freeze
  • Spectral gating, freezing, blurring, shifting and stretching
  • Loop any part on a player that keeps its place when the sound changes
AudioGranularAudioFreezeFrameAudioCrossSynthAudioSpectralProcessAudioClipPlayer
Turn a clip into a cloud →Freeze a moment into a drone →Loop a room tone →
a keyboard synth on its own panel: waveform, cutoff, attack and release, with a scope, a spectrum and a level meter
a synth on a panel you lay out
05Synthesis

Make sounds from nothing.

Noise, oscillators, FM, plucked strings and struck metal, through filters and envelopes you lay out on a panel of your own. Your CLAP and VST3 synths and effects sit in the same chain, each in its own process if you like.

  • A modal synth for bells, bars and metal hits, struck again on any trigger
  • Plucked strings, waveguides, and harmonics you draw by hand
  • Move a knob on your MIDI controller to map it to any setting
AudioNoiseAudioFMOperatorAudioModalSynthAudioKarplusStrongPluginHost
Tutorial: patch a synth you can play →Build synths →Run a plugin you do not trust →
the Dynamics menu open: compressor, limiter, noise gate, expander, de-esser, transient shaper, multiband compressor and duck under a take
the Dynamics menu
06Deliver

Mix it and deliver to spec.

De-ess and compress the dialogue, duck the music under a voice, then set the loudness to -16, -14 or -23 LUFS and save WAV, AIFF, FLAC, MP3 or OGG with the tags written in. Find every hit in a take, split at each one and save each part as its own file.

Dynamics›De-esser→Dynamics›Duck under a take→Level›Loudness›Normalize to LUFS→Send to›Save to a sound file
  • A meter for momentary, short-term and integrated loudness and true peak
  • Copy a set of edits onto every open sound at once
  • Run the same chain over a whole folder from the command line
AudioDeEsserAudioDuckingAudioLoudnessAudioNormalizeOutputAudioFile
Tutorial: tidy a podcast episode →Tidy a podcast →
+Every job

Every edit stays open.

  • A step you can reopenEach pick from the menus adds one step to the Stack and one node to the canvas. Click any step to hear the sound at that point and change it.
  • Before and afterCompare plays a step against the one before it at the same moment, or the first sound against the last.
  • Recipes and versionsKeep a set of steps as a recipe for the next sound, or branch a second version from any step and keep the first.
  • No render to wait forEdit with the graph stopped and press Listen. Long sounds work in the background and say so.
Sound under picture

Lay sound under the picture.

The video timeline has picture tracks and sound tracks, so effects sit on the cut and render out with it. The fire, the flood and the falling wall come from the same app, so you can make the shot and its sound side by side.

the video timeline: nine simulation renders on the picture track, and impacts, whooshes and a room tone bed on three sound tracks below, with a whoosh selected and its volume, pan and fades open on the right
nine simulations from Kinematic, cut over impacts, whooshes and a room tone
  • Sound tracks under the picture, with mute and solo on each track and volume, pan and fades on each clip
    VideoArrangementPlayer
  • Render the cut with its mix as AAC, MP3, FLAC, Opus or PCM sound
    OutputVideoFile
  • Set a sound file to follow the project timeline beside a simulation, then render the range with the sound in it
    InputAudioFile
  • Read LTC off an audio track, or MIDI timecode, and lock to the picture
    MidiLtcDecodeMidiTimecodeIn
a wall shatter, ready for its impact
a dam break, ready for its roar
In the Demos browserFireballA fireball that ignites in one burst, blooms and fades away: a shot to score.
Edit video →Tutorial: cut a short video →Simulate →
Sound into pictures

Let sound move the picture.

The level of a sound, each hit and each band can drive any number in a scene. A voice swells a shape, a hit throws particles, the low end lifts the ground. Save it as a video with your sound in it.

  • Follow the level, find every hit, and split the sound into lows, mids and highs
    AudioEnvelopeFollowerAudioOnsetDetectAudioCrossover
  • Drag any of them onto a setting in the scene to drive it
  • Draw a sound as a waveform, a spectrum, a spectrogram or a vectorscope
    AudioToImage
In the Demos browserAudio danceA figure dancing to a track: the level of the music swings its spine and arms.
Sound to picture →Turn sound into numbers →Tutorial: a visual that pulses →
a shape swelling and twisting with a voice
particles driven by a song
a box pulsing with the loudness
Sound in the room

Put sound in a room or a show.

Place a sound in 3D and let it fall away with distance, give it a room, or encode it to ambisonics and decode to stereo or four speakers. Fire it from the show, a camera or a sensor, and leave it running for a season with no window open.

  • Spatial pan, a room simulator, and first-order ambisonics in and out
    AudioSpatialPanAudioRoomSimAudioAmbisonicEncodeAudioAmbisonicDecode
  • Take cues from QLab over OSC, from MIDI Show Control, MIDI or a WebSocket
    OSCReceiveMidiShowControlInWebSocketServerInterface
  • A hand gesture can fire a sound, worked out on your own machine
    LocalAIImageToHandGestures
  • Echo cancel keeps an installation from hearing its own speakers, so it hears the visitor
    LocalAIAudioEchoCancel
  • Edit sounds while the show runs. Edits to a long sound wait until you press Send
In the Demos browserSpatial audio orbiterA tone circling the listener, with distance and early reflections from a room.needs audio device
Fly a sound around your head →Build a sound cue for a show →Run a live show →
an immersive room, for sound and picture
a tone circling the listener: the left and right channels trade places as it moves
a sound moving from left to right
an open palm read from a webcam, ready to fire a cue
an open hand, read from a webcam
How it fits

Your files, plugins and gear.

Bring your recordings, your plugins and your controllers as they are. The full list is on the Works with page.

Files
Opens
WAVFLACMP3OGGAACMP4, MOV, MKV, WebM and AVI with soundMIDI files
Saves
WAVAIFFFLACMP3OGGVideo with soundMIDI files

Sound files save with title, artist and the other tags, at the rate you ask for up to 96 kHz. A long take records crash-safe, so a power cut leaves a file you can play, and a WAV past 4 GB stays one file.

Audio interfaces and mics
Record from any input, and play out to your speakers.
MIDI keyboards and controllers
Play the synths, and move a knob to map it to any setting.
Audio plugins
CLAP and VST3 on every platform, LV2 on Linux. Save presets, and run any plugin in its own process.
The cut to work against
Open an MP4 or MOV with its sound and lay your effects under it on the video timeline.
Your DAW
Kinematic runs beside it. Follow its MIDI clock, or play into it through a virtual MIDI port on macOS and Linux.
Shows, sync and games
  • QLab and show controlOSC over UDP or TCP, MIDI Show Control and MMC, in and out.
    OSCReceiveMidiShowControlOut
  • TimecodeLTC read off an audio track, MIDI timecode in and out, and MIDI clock.
    LtcDecoderMidiTimecodeOut
  • Games and web pagesA WebSocket server or client, and HTTP requests, so a game or a page can call for a sound.
    WebSocketClientInterfaceHTTPRequest
  • Ableton LiveTempo, clips and scenes, through the free AbletonOSC script.
    AbletonLiveLink
A folder at a time

Build the chain once on one sound and save the graph. Then run it over every file in a folder, four at a time, or make four pitches of one whoosh, with no window open.

terminal
$ kinematic run clean.kin --each in.file_path='raw/*.wav' \
    --set save.file_path='clean/{stem}.wav' --for 30s --jobs 4
$ kinematic run whoosh.kin --sweep pitch.semitones=-5,-3,0,2 \
    --set save.file_path='whoosh_{i}.wav' --for 20s
Coming from Reaktor?Synths, panels and macros, and what they are called here → Building installations in TouchDesigner?How the same ideas carry over → Everything it works withFiles, protocols, gear and apps, in and out →
Where to start

Tutorials and demos.

Each tutorial stands alone and takes a few minutes. The demos come with the app: open Help, then Browse Demos, and take one apart.

Tutorials
8 min · First stepsClean up a field recordingThe steady noise turned down, the rumble gone, the bumps trimmed and a fade at the end.Read the tutorial →8 min · First stepsTurn a creak into a whooshA swelling whoosh for a film or a game, from any short sound.Read the tutorial →8 min · First stepsMake a voice sound like an old radioA voice turned thin, crunchy and boxed in, saved as a sound file.Read the tutorial →12 min · Getting goingTidy a podcast episodePauses cut, S sounds softened, the level evened out and set to -16 LUFS.Read the tutorial →12 min · First stepsPatch a synth you can playA saw, a filter and an envelope, played from keys on screen, with a panel of four knobs.Read the tutorial →12 min · First stepsCut a short videoYour clips in order and trimmed, with fades and a title, saved as an MP4.Read the tutorial →
Demos to open first
Audio deviceSound design studioA wavetable editor, a spectrum sculptor, an EQ designer, an LFO curve and a mixer on one panel.In the Demos browserWaveshaper chainA tone pushed through a waveshaper, a saturator and a wave folder, each adding its own grit.In the Demos browserSVF filter sweepNoise through one filter, heard as low pass, band pass, high pass and notch at once.In the Demos browserFlanger + phaser demoA flanger and a phaser side by side on the same sound, beside the dry one.In the Demos browserADSR envelope demoPress a toggle to strike a note and hear attack, decay, sustain and release.In the Demos browserStep sequencer drum machineKick, snare, hats and toms built from oscillators and noise, on a step grid.In the Demos browser
All the Play tutorials and guides →Edit sound →Browse every demo →
Honest gaps

What it does not yet do.

Better to know now. Something missing that you need? Reach out.

  • More than 16 bits in an editSound travels between nodes at 16 bits. Each node works in float inside, but a 24-bit or 32-bit float save still holds 16-bit sound.
  • Spectral repairThere is no spectrogram to paint out a cough, a bird or a siren. Denoise handles steady noise: hiss, hum and room tone.
  • Surround and binauralNo 5.1 or 7.1 panner and no HRTF headphone mix yet. Spatial pan is stereo with distance, and the ambisonic decoder gives stereo or quad.
  • Audio Unit pluginsThey do not load. CLAP and VST3 do on every platform, and LV2 on Linux.
  • Sessions from your DAWIt does not open AAF or OMF. Bring the sounds in as files.
  • Long recordings in the editorA clip edits up to 30 minutes of stereo. A longer file still plays, but not as a clip you can edit.
  • Game audio middlewareNo Wwise or FMOD banks. A game can call for a sound over OSC or a WebSocket instead.
Your next sound

Download Kinematic.

Free. No account, no license key, no subscription, even for commercial use. macOS, Windows, Linux and Raspberry Pi, on every machine in the studio.

Download for macOS 0.5.0Reach out
ground raised by the lows, mids and highs of a song