Jackdaw Audio Perception Lab

Binaural Beats Lab

Headphone-first tone engine for sleep, focus, meditation, and cognitive flow.

Ready

Session Controls

Build your own tone stack or apply a curated preset, then start playback.

220 Hz
00 kHz 220 Hz . 0
kHz
Hz
0.1
Segment steps: 1000 Hz, 1 Hz, and 0.1 Hz
Sets the overall loudness. Keep it modest to avoid ear fatigue over longer sessions.
45%
Pink noise adds a soft broadband bed that can make tones feel less stark and easier to listen to.
0%
Brown noise is deeper and heavier than pink noise. Useful for grounding and relaxation sessions.
0%
7.0 Hz (Theta)
07.0 Hz
1 Hz
0.1 Hz
Range: 0.5 to 42.0 Hz
Delta · 0.5-4 HzDeep restoration and slow-wave sleep support.
Theta · 4-8 HzReflective, meditative drift and decompression.
Alpha · 8-14 HzRelaxed alertness with steady cognitive tone.
Beta · 14-30 HzTask-drive, concentration, and active thinking.
Gamma · 30-42 HzHigh-intensity focus and mental integration.
Selects the tone shape. Sine is smoothest; the brighter waves can sound sharper and more energetic.
Chooses how the beat is produced: binaural (split ears), monaural (mixed), or isochronic pulses.
Automatically ends playback after the selected duration. Handy for naps and focus blocks.
Applies staged changes for a specific outcome, such as sleep onset or deep work priming.
Gently fades the session near the end, so the finish feels natural rather than abrupt.
Plays a short chime when the timer completes. Useful if you’re timing focused work.
Choose an ambience texture and dial in the mix to sit underneath the tones without overpowering them.
0%
Protocol Builder
Stages run in order using stage minutes and target values. Start with 2-4 stages. Delta controls beat spacing, volume is final stage gain, and noise layers are mix beds.
Inactive
Stages use current base frequency and move delta/noise/volume over each stage duration.
Left Ear 220.0 Hz
Right Ear 227.0 Hz
Preset Custom
Session Phase Static

Preset Library

Evidence Level: Mixed

Citations: Preset-level references shown after selection.

Stereo Check

Use headphones and verify channels before a session.

Psychoacousticssuite

—

Partials left = f1, right = f2. Height is frequency.

XY scope x = tone 1, y = tone 2

Local minima the scale this timbre wants — click a row to jump there

f2centsratioerrorroughness

Roughness = Σ over every pair of partials of Plomp & Levelt's dissonance curve (Sethares' parameterisation, min(a₁,a₂)·(e−3.5x − e−5.75x), x = 0.24·Δf/(0.0207·fmin + 18.96)). Amplitudes are normalised per tone, so 0 is perfectly smooth and ~0.18 is the maximum a single partial pair can contribute.

—

Below ~15 Hz you hear one tone throbbing. Above that it turns to roughness, and once Δf exceeds the critical bandwidth the two tones separate out and the roughness dies away. Pan them to opposite ears and the roughness mostly vanishes — it is generated inside one cochlear filter, not in the air.

—

Drag the lowest harmonic up past 1 — the fundamental disappears from the spectrum but the pitch does not move. Now add a constant shift to every partial: the spacing stays the same, yet the pitch slides. That is the pitch of the residue, and it is why the model is pattern-matching, not a filter.

—

Octave-spaced sines under a fixed bell-shaped loudness envelope. Every component rises, but the envelope never moves, so the spectrum after one octave is identical to where it started. Pitch chroma climbs forever; pitch height goes nowhere.

—

Band-limited noise masker plus a pulsed probe. Drop the probe until it vanishes, then move it away in frequency and it reappears at the same level. The shaded region is the predicted masked threshold — an ERB-wide filter skirt, shallow upward in frequency and steep downward, which is why bass masks treble far better than the reverse.