Project Latest Update: 26 July 2024
The stress pattern was based on The CMU Pronouncing Dictionary.
The cmudict module from the NLTK library was used to extract the stress pattern from the dataset.
The English words dataset was based on the SubtlexUS dataset.
According to what is mentioned on the CMU Pronouncing Dictionary website, "Stress is difficult to get right and people disagree about it."
Visualizations Latest Update: 6 May 2024
- Execute
main.py
- Data is loaded to a local SQLite database automatically.
- A local SQLite database is created automatically if not exist in the given path.
- You can change the path by adjusting this variable:
db = 'eng_stress_pattern.db'
- Data is loaded to a local SQLite database automatically.
eng_stress_pattern_finder.py
- Find the stress pattern of the English words with the given dataset.
extract_data.py
- Extract data from SubtlexUS dataset.
transform_data.py
- Transform data to find a syllable count and stress pattern of each English word.
- Words that aren't in the dictionary will be filtered out.
load_to_sqlite.py
- Load data to SQLite database tables.
Tables:
- StressPattern
- Store syllable count, stress pattern, and primary and secondary stress position of each word
The CMU Pronouncing Dictionary: http://www.speech.cs.cmu.edu/cgi-bin/cmudict
“SubtlexUS” dataset: http://www.lexique.org/?page_id=241