This is the repository for the US English formant normalization paper, evaluating normalization accounts against US English perception data.
- data/: participant data, stimuli formant data, phonetic statistics for the vowel database used in modeling.
- experiments/: scripts and stimuli for the two experiments.
- models/: models created by scripts.
- output/: files for generating the Persson, Barreda & Jaeger 2025 paper, including submitted versions.
- scripts/: R and Praat scripts for generating the synthesized stimuli, analysing formant data and repairing/correcting automatically extracted formants.