Skip to audio lab
← Audio & Acoustics

The audio lab

Change it. See it. Hear it.

Three hands-on experiments in signal processing and acoustics.

SYNTHETIC SIGNALS

Press a listen button to play a four-second sample. No microphone or uploads needed.

01

Signal processing

Find the signal in the noise.

A repeating musical tone, mixed with broadband hiss. How much can a simple filter help?

Shape the sound

Lower values mean more noise.

Lower cutoffs remove more hiss—and can dull the wanted signal.

What passes through?

Filter response
Low-pass filter frequency responseThe filter increasingly attenuates frequencies above the selected cutoff.
Input SNR
Output SNR
Wanted signal retained

Output SNR compares filtered signal power with filtered noise power. Signal retained shows output signal power relative to the clean input; listen for distortion as well as noise reduction.

Where this matters

Sensor conditioning, speech front ends and embedded audio. Filtering is a useful baseline; separating overlapping speech and noise calls for more advanced methods.

What this experiment models

A deterministic four-second harmonic melody at 24 kHz, with white noise scaled to the chosen input SNR, processed by a second-order Butterworth low-pass filter. The clean and noise components are filtered separately to calculate the displayed SNR. Playback uses the same gain across comparisons; there is no per-sample loudness normalization. This is classical DSP, not a trained speech-enhancement model.

02

Spatial acoustics

Point the listening beam.

Two tones arrive from different directions. Steer a virtual microphone array toward the sound you want.

Control the array

Microphone spacing: 4 cm. More microphones increase the aperture and sharpen directional selectivity.

A / −35°1,000 Hz tone
B / +45°650 Hz tone

Directional sensitivity

0° = broadside
Microphone array directional responseTwo frequency response curves show sensitivity across arrival angles; vertical markers locate sources A and B.
Blue: response at 1,000 HzAmber: response at 650 Hz
Source A gain
Source B gain
Array aperture

0 dB means fully retained; negative values mean attenuation. Directional response depends on frequency. Both sources can remain audible.

Where this matters

Microphone arrays, directional sensing and audio capture on devices. Geometry, frequency, reverberation and compute budgets shape real-world performance.

What this experiment models

An ideal uniform linear delay-and-sum array in the far field, with sound speed 343 m/s and equal source amplitudes. The chart uses the normalized complex array factor. Playback applies that factor’s gain and phase to each synthetic tone. It models two steady tones, not speech, moving sources, reverberation, microphone mismatch or an actual hardware recording.

03

Active noise control / ANC

Can sound cancel sound?

Match an opposing tone to a 180 Hz hum. Adjust its phase and amplitude to find the quiet point.

Tune the opposing sound

180° opposes the original tone; 0° reinforces it.

100% matches the amplitude of the original hum.

Two waves. One result.

At one observation point
Original, opposing and combined waveformsAdjust phase and amplitude to change the combined waveform.
Blue: originalAmber: opposing toneCyan: combined result
Level change
Residual amplitude
Noise frequency180 Hz

Negative dB means attenuation; positive dB means reinforcement. Exact cancellation produces silence in this ideal simulation.

Why the engineering matters

Real ANC depends on microphone and speaker placement, acoustic transfer paths, timing and stability. Cancellation at one point does not imply a quiet room; moving the listener or changing the sound changes the problem.

What this experiment models

Linear superposition of two steady 180 Hz tones at one observation point. Residual amplitude relative to the original is the magnitude of 1 + a·exp(jφ), where a is the amplitude ratio and φ is the phase offset. Playback uses the actual summed waveform with the same gain as the original. The phase and amplitude are set manually: this illustrates the cancellation principle, not a feedback controller, adaptive ANC algorithm or measured headphone performance. Graphs use normalized amplitude. All sounds are synthesized in the browser.

Discuss your audio or edge-AI project →Explore all demonstrations

MLAIA DATA SCIENCE

Tell us what
you’re working on.

Share the problem, the data you have and what success would look like. We’ll discuss a practical next step.

Please don’t include confidential datasets, credentials or patient information.

yochai@mlaia.com
+972 52 484 6282

Your inquiry is handled under our Privacy Policy.