Animating with Audio in Blender: Sound to Samples Guide

Having a cube bounce with every drum hit, or lights flashing on the drop, does not require frame-by-frame animation: the Sound to Samples workflow in Blender turns your track into a curve that drives your objects. Here is the full pipeline, from importing the audio to beat-syncing.

What Animating with Audio in Blender Is

From sound to animation curve

The core of this workflow is the Sound to Samples operator in the Graph Editor: it creates a sampled curve based on a sound file. That curve reflects the amplitude of the audio over time and, like any other f-curve, can drive any animatable property in the scene: the scale of an object, the strength of a light or the position of the camera.

The operator is not new: in older versions it was called Bake Sound to F-Curves and was renamed during the Blender 4.0 cycle. The old name is still useful as a search term, but in current menus you will find it as Sound to Samples, under the Channel menu of the Graph Editor.

When to use it (and when not to)

It is the ideal tool for audio-reactive effects: visualizers, lights pulsing to the beat or objects bouncing with the music. It also works as an automatic sync skeleton when you animate with voice: the amplitude curve gives you a starting point that you later refine by hand. It does not replace fine pose-to-pose animation: the automatic curve is the beginning of the work, not the final result.

Setting Up Audio in Blender

Importing your sound track

Audio is added as a Sound Strip in the Video Sequencer with the Add menu, Sound option. Strips reference the external file instead of embedding it, so if you edit the track in an external audio editor, the changes show up when you reload the file. You can stack several strips as independent tracks (dialogue, music, ambience, effects) and they mix together during playback and rendering: Blender works as a multitrack mixer.

If the file contains several audio streams, Blender creates one strip per stream, automatically synced with the source file timing. Common formats go through FFmpeg: WAV, MP3, OGG, FLAC and AAC.

Viewing the waveform

To work with sound visually, enable the waveform: the Sequencer overlay can show all waveforms, none, or follow the per-strip option, and each strip also has its own Display Waveform setting. A useful detail: clipping, meaning values above 100 percent amplitude, is painted red in the waveform, warning you about peaks that can distort the analysis.

Listening in sync

The Playback menu in the Timeline controls how audio behaves during playback. No Sync tries to play every frame no matter what, which can desync the audio; Frame Dropping drops viewport frames to match the real frame rate; and AV-sync syncs playback with the audio clock, dropping frames if the view runs slow. When animating with sound, AV-sync is the recommended mode: what you see and what you hear stay together.

Sound to Samples: From Audio to Curve

You need an existing curve first

The operator has a requirement that confuses many users: it needs at least one selected f-curve to activate. If it appears greyed out, no curve is selected. The correct flow is to keyframe the property you want to move first (for example, the Z scale of a cube with an initial keyframe), open the Graph Editor and select that curve.

Running the operator

With the curve selected, go to the Channel menu, Sound to Samples option, in the Graph Editor. Blender asks for the audio file and generates the sampled curve over the selected one. The result is dense samples rather than traditional keyframes: the curve is ready to drive the property to the rhythm of the sound.

Tuning the audio analysis

The operator exposes parameters verified in the official Blender manual. Lowest Frequency and Highest Frequency are the cutoff frequencies of a high-pass and a low-pass filter: they let you isolate a band of the spectrum, such as the bass of the drums or the highs of a synthesizer. Attack Time and Release Time control the hull curve, meaning how fast the curve can rise and fall: lower values react more sharply. Threshold is the minimum amplitude needed to influence the curve, and the output modes define how amplitude is translated: Accumulate only sums positive amplitude differences, Additive adds the amplitudes together, Square outputs a square curve (negatives to -1, positives to 1) and Square Threshold sends everything below the threshold to 0. The full reference is in the Graph Editor documentation.

From Sampled Curve to Animation

Samples to Keys

If you want to edit the curve by hand, the Channel menu, Samples to Keys option, converts it into editable keyframes: it creates one keyframe on every frame. It is useful for cleaning peaks, smoothing passages or accenting specific moments; Keys to Samples does the reverse.

Using the curve as a driver

The sampled curve can be copied and pasted as a driver into other properties or other objects. This is the classic visualizer workflow: a single audio analysis moves dozens of objects at once, all reacting the same way.

Controlling the intensity

The amplitude of the curve rarely fills the full range of the property you are animating. It is worth remapping it, with a driver scale factor or with scale and offset keyframes on the property, so the motion is clearly visible without reaching extreme values.

Beat Syncing: Markers and BPM

Calculating the beat interval

If you know the BPM of the track, you can find the strong beats without listening frame by frame. The interval in frames between beats is 60 divided by the BPM, times the frame rate: at 120 BPM and 24 fps that is 12 frames per beat; at 30 fps it is 15; at 90 BPM and 24 fps it is 16.

Marking beats in the timeline

With that interval calculated, add markers in the Timeline with the M key on the strong beats and place keyframes on those markers. During playback, the playhead jumps from beat to beat and the animation stays visually anchored to the music.

Combining automatic and manual work

The best results are usually hybrid: use Sound to Samples for the base motion (the bounce, the flashing) and adjust poses or accents by hand on the markers. Mixing automatic curves with human judgment is what separates a generic effect from an animation that truly follows the music.

Practical Use Cases

Audio visualizers

The most popular case: a group of cubes, spheres or bars scaling with the frequency band you choose. To speed things up, the free Simple Audio Visualizer add-on is available on the official Blender extensions platform: you select the objects, pick the audio file, adjust the intensity multiplier and choose the transform channels to animate.

Lights and camera reacting to music

The same curve can animate the strength of a light or the position of the camera. For lights, keyframe the emission strength or the light power and apply Sound to Samples to that curve: the intensity changes will follow the audio contour.

Voice-over and basic lip sync

With a voice-over, the amplitude curve can guide the mouth opening using shape keys: the mouth opens with voice peaks and closes in the pauses. It is an automatic skeleton that you then polish by hand; if you want to go deeper into facial expression, the blog has specific facial animation material.

Common Mistakes and How to Fix Them

The curve does not react

Check three things: that an f-curve is selected (keyframe the property first), that the chosen frequency band matches the audio content (try the full range and then isolate) and that the threshold is not so high that almost nothing passes the minimum amplitude.

The motion feels noisy

Attack and release values that are too low make the curve follow every micro-variation of the audio. Raise the release time or narrow the frequency band to smooth the result.

Audio does not play back

Check the sync mode in the Playback menu (AV-sync), that the strip is not muted or on a hidden channel, and that the file format is decodable by Blender through FFmpeg.

Conclusion

Animating with audio in Blender turns a tedious synchronization problem into an automatic workflow that takes minutes: import the track, keyframe a property, run Sound to Samples, tune frequencies and beats, and the result is a scene that dances with its soundtrack. The automatic curve does not replace judgment: it amplifies it. If you want to see the workflow in action and learn with step-by-step examples, explore the curated animation video tutorials on the platform.