Skip to content

About

I built an experiment that turns sound into real-time interactive 3D plots. Using STFT and spectral features centroid, spread, and flatness, the audio is mapped into a 3D space with Plotly, spynced to an HTML audio, player. Apllications in bioacoustic example birdsong, noise analys, live music visuals, and data science.

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

8 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 

Repository files navigation

Exploring-Audio-with-3D-Visualization-in-Python

Exploring Audio with 3D Visualization in Python

In recent days I’ve been experimenting with an approach to translate sounds into visual data structures, creating interactive plots that draw themselves in real time as the audio is played.

Technical flow:

Audio preprocessing (normalization and conversion to mono).

Spectral analysis with STFT (Short-Time Fourier Transform) to capture the evolution of frequencies over time.

Extraction of spectral features:

Spectral Centroid → the “center of mass” of the spectrum (related to the “brightness” of the sound).

Spectral Spread → dispersion around the centroid (how spread out the frequencies are).

Spectral Flatness → measure of tonality vs. noisiness (tonal sounds ≈ 0, noise ≈ 1).

Interactive 3D rendering with Plotly, where:

X-axis = time (s)

Y-axis = centroid (Hz)

Z-axis = spread (Hz)

Color = flatness (heatmap)

Integration with an player in HTML, so that the visualization is drawn in sync with the playback.

Practical applications:

Bioacoustics: analyzing bird songs and acoustic biodiversity

Audio engineering: noise characterization (e.g., engines, turbines, airplanes)

Music: live artistic visualizations during performances

Data science: exploring time–frequency relationships in complex signals

Tech stack: Python · NumPy · SciPy · Plotly · Google Colab

The result is an interactive, synchronized 3D graph that transforms any audio into a dynamic visual experience bridging science, technology, and art.

Birdsong 3D Visualization

Here’s an example of the visualization:

Plot

About

I built an experiment that turns sound into real-time interactive 3D plots. Using STFT and spectral features centroid, spread, and flatness, the audio is mapped into a 3D space with Plotly, spynced to an HTML audio, player. Apllications in bioacoustic example birdsong, noise analys, live music visuals, and data science.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages