ruspeechstress

this work is based on the code presented in (Bernhard et al. 2022).

requirements:

• 6.1.39 <= Praat => 6.2.09

• 3.8 <= Python => 3.10

• R == 4.3.0

• ffmpeg == 6.0

• rusyllab 0.0.4

• requirements listed in reqirements.txt

usage

pip install ruspeechstress

from ruspeechstress import create_dataset, train_model, extend_dictionary

create_dataset(dataset_dir="path to .wav and .txt files", dictionary_path="path to dictionary.txt")
train_model(dataset_dir="path to .wav and .txt files", features path="path to .tsv-features collected on previous step", clf='name of sklearn classifier')
extend_dictionary(dataset_dir="path to .wav and .txt files", dictionary_path="path to dictionary.txt", classifier='path to .pkl classifier', scaler='path to .pkl scaler')