AURORA: Acoustic Understanding and Real-time Observation of Resonant Articulations
AURORA provides real-time visualization of spectral features with formant tracking and (preliminary) tongue shape estimation. The application captures audio input, performs spectral analysis, and displays spectral peaks and predicted tongue shape in real-time. An example screenshot of the GUI is shown below:
- Python 3.12 or higher
- uv (for dependency management)
If you don't have uv installed, follow the installation instructions at https://docs.astral.sh/uv/getting-started/installation/.
# Navigate to the top-level directory
cd aurora
# Install dependencies
uv syncRun the main application:
uv run python aurora/aurora.pyThe config.yaml file allows you to select between different models, rotate the position of the static vocal tract contours, and set different default acoustic settings. The defaults are:
audio:
fs: 10000 # sampling rate
frame_size: 1024 # frame size
lpc_order: 12 # tune LPC order for speaker-specific formant tracking
rms_threshold: 0.03 # only track acoustic input above threshold
data:
tongue_model: data/tongue_model.pkl # model for formant-tongue inversion
template: data/aurora_template.npz # template for static vocal tract contours
contours:
rotation_deg: 0.0 # angle to rotate static contours (in degrees)
Sam Kirkham (s.kirkham@lancaster.ac.uk)