pyvslm — Virtual Sound Level Meter

A Python implementation of a virtual sound level meter (VSLM) for post-processing digital audio recordings as if they were measured by an ANSI-compliant sound level meter. Supports A, C, and Z frequency weightings; Fast, Slow, and Impulse time weightings; octave and 1/3-octave band analysis; PSD; and spectrogram modes.

---

Requirements

• Miniforge or Anaconda with the conda-forge channel
• Windows, macOS, or Linux

---

Environment Setup

conda create -n pyvslm -c conda-forge \
    python=3.12 numpy scipy matplotlib pyside6 \
    pysoundfile python-sounddevice numba \
    pydantic pyyaml pyinstaller pytest

conda activate pyvslm

---

Running the Application

conda activate vslm_env
python run_pyvslm.py

The GUI will open. Load a WAV file with the Load WAV button, optionally select a sub-section with Select Section, set your analysis mode and parameters, then click ANALYSE.

---

Running the Tests

Unit tests

conda activate vslm_env

# Run all unit tests
python -m pytest tests/ -v

# Run a single test file
python -m pytest tests/test_engine.py -v
python -m pytest tests/test_filters.py -v

# Run a specific test
python -m pytest tests/test_engine.py::TestBroadbandLeq::test_steady_state_leq

Filter visualisation plots

These scripts display matplotlib figures — useful for visually verifying filter compliance.

# A and C weighting responses vs IEC 61672-1 ideal curve + Class 1 tolerance band
python tests/plot_weighting_filters.py

# 2×2 grid: magnitude and error (digital − analogue) for all sample rates,
# with both IEC 61672-1 Class 1 and ANSI S1.4 Type 0 tolerance overlays
python tests/plot_weighting_filters_compact.py

# Octave filter bank response vs IEC 61260-1 / ANSI S1.11 Class 1 spectral mask
python tests/plot_ansi_filters.py

# 1/3-octave version of the above
python tests/plot_ansi_filters.py third

---

Analysis Modes

Mode	Description
Level vs Time (Lp)	Time-weighted instantaneous SPL history
Leq Analysis	Equivalent continuous level with noise dose
Octave Bands	1-octave IIR filter bank (ANSI S1.11)
1/3 Octave Bands	1/3-octave IIR filter bank (ANSI S1.11)
PSD	Power spectral density via Welch's method
Spectrogram	Short-time PSD vs time

---

Project Structure

run_pyvslm.py          Application entry point
vslm/
├── constants.py       Enumerations, interval map, mode map
├── settings.py        Pydantic settings model, load/save
├── controller.py      MVC controller
├── dsp/
│   ├── engine.py      StreamProcessor — core analysis engine
│   ├── calibration.py Acoustic calibration utilities
│   ├── leq.py         Leq statistics and noise dose
│   ├── exporter.py    CSV export
│   └── filters/
│       ├── weighting_filters.py   IEC 61672-1 A/C/Z filters
│       └── octave_filters.py      Octave / 1/3-octave filter banks
└── gui/
    ├── main_window.py
    ├── worker.py
    ├── plot_widget.py
    ├── plot_manager.py
    └── dialogs/
        ├── calibration.py
        ├── waveform.py
        └── about.py
tests/
├── test_engine.py                  StreamProcessor unit tests
├── test_filters.py                 Filter compliance tests (IEC 61672-1 Class 1)
├── plot_weighting_filters.py       A/C weighting response plots
├── plot_weighting_filters_compact.py  2×2 magnitude + error grid
└── plot_ansi_filters.py            Octave / 1/3-octave bank compliance plots

---

Settings

Application settings are saved automatically to settings.yaml on exit and restored on next launch. You can also save and load named settings files via File → Settings.