Metriq Visualizer v1.13.0

Metriq Visualizer is an open-source multidimensional data and media visualizer.

It turns local audio, video, CSV, TSV, and XLSX data into interactive 3D geometry that you can scrub, inspect, and export as an MP4.

About Metriq

Metriq builds technology intended to support real-world progress. The visual language behind this project reflects that broader design philosophy: clean, technical, and forward-looking.

Features

• Local audio and video analysis
• Local CSV, TSV, and XLSX import for numeric datasets
• Formula-based mapping for X, Y, Z, color, and size
• Interactive geometry playback and timeline scrubbing
• Optional smooth spline path rendering for line and tube modes
• Save and reopen visualizer projects as .mvproj
• Save and reuse visualizer presets as .mvpreset
• Visualizer behavior presets are loaded from the local presets/ directory next to the app. Add your own .mvpreset files there to populate the list.
• Legacy .bgl project files can still be opened
• Feature reference panel and mapped trace panels
• Professional dark-mode interface
• MP4 export presets in:
  • 1280×720 landscape
  • 1920×1080 landscape
  • 1080×1920 vertical
• Export engine selector with Auto GPU encoder → CPU fallback, CPU FFmpeg, and legacy OpenCV modes
• Fast, cached audio analysis (about 6x faster than v1.10.18 with identical results)
• Safe saves and exports: files are written to a temporary file and moved into place only on success
• Stage Output (View menu): mirror the live viewport and panels to a second display for performances
• Export mapped data as CSV or NPZ (File menu)
• Verified updates (Help menu): checks GitHub releases, verifies a SHA-256 checksum, and installs only after you confirm

Input notes

Open a media file to extract a feature set and build geometry.

Open a table file with at least one numeric column to map imported values into geometry. Imported columns are available as formula-ready features, and the table importer also derives helper features such as pc1, pc2, pc3, magnitude, column_mean, and delta_magnitude.

Formula examples

For media:

• pc1
• 0.7*mfcc_1 + 0.3*chroma_mean
• smooth(spectral_flux, 5)

For tabular data:

• input_1
• mean(input_1, input_2)
• pc1
• delta_magnitude

Supported functions: abs, sqrt, log, log1p, exp, clip, smooth, mean, avg, sum, max, min

Install from source (Linux)

Requirements:

• Python 3.10+
• FFmpeg

Ubuntu / Debian:

sudo apt update
sudo apt install ffmpeg python3 python3-pip python3-venv \
    libgl1 libegl1 libxkbcommon-x11-0 libxcb-cursor0 libpulse0

Option 1 — Clone from GitHub

git clone https://github.com/MetriqOrg/Metriq-Visualizer.git
cd Metriq-Visualizer

python3 -m venv .venv
source .venv/bin/activate

pip install -r requirements.txt

python metriq_visualizer_app.py

Option 2 — Download ZIP from GitHub

GitHub source ZIP files usually extract to a folder named Metriq-Visualizer-main or Metriq-Visualizer-master, depending on the default branch. Use the actual folder name that was extracted.

# Example for GitHub's default main-branch ZIP name:
cd Metriq-Visualizer-main

python3 -m venv .venv
source .venv/bin/activate

pip install -r requirements.txt

python metriq_visualizer_app.py

Option 3 — Launcher script

From inside the project folder:

chmod +x run_linux.sh
./run_linux.sh

Install from source (macOS)

For additional macOS compatibility and packaging notes, see docs/MACOS.md.

Requirements:

• Python 3.10+
• Homebrew
• FFmpeg

Install dependencies:

brew install python ffmpeg

Clone the repository:

git clone https://github.com/MetriqOrg/Metriq-Visualizer.git
cd Metriq-Visualizer

Create a virtual environment:

python3 -m venv .venv
source .venv/bin/activate

Install Python dependencies:

pip install -r requirements.txt

Launch:

python metriq_visualizer_app.py

If macOS blocks the app or Python process because it was downloaded from the internet, open System Settings → Privacy & Security and allow it there. For source installs, this is usually less of an issue than with packaged .app builds.

Packaged builds

For normal users, packaged releases are recommended over source installs. Keep compiled binaries out of the repository and attach them to GitHub Releases instead.

Recommended release assets:

• Windows ZIP or EXE
• macOS Apple Silicon ZIP or DMG
• macOS Intel ZIP or DMG
• Linux AppImage or ZIP

See docs/RELEASE_BUILDS.md for release-build notes.

License

The source code in this package is licensed under MPL 2.0. See LICENSE.

Branding and intellectual property notice

Use of this software, any fork, any modified version, or any derivative work does not grant permission to use the Metriq name, trademarks, service marks, logos, symbols, trade dress, copyrighted brand materials, or any other Metriq Foundation, Inc. intellectual property, and does not imply affiliation, sponsorship, or endorsement by Metriq Foundation, Inc.

See TRADEMARKS.md and assets/ASSET_NOTICE.md for the brand-asset reservation notice.

Copyright

Copyright (c) 2026 Metriq Foundation, Inc.