Streamlit app running Python code for comparative analysis of International Phonetic Alphabet (IPA) strings. App is constructed to streamline data exploration whilst deciding what directions to take in research. Repository also includes data and research from a Spanish phonetic acquisition pilot study from 2020.
The intention of this repo is to perform contrastive data analysis between prescriptive and descriptive IPA strings.
I have made all of the data I collected for this study open-source.
The bulk of the data collected is in the transcriptions/descriptive/ folder. Here there are a number of files named particX.txt; each doc contains a long string of the participant's recording (their pronunciation of the wordlist) transcribed into IPA. I did all of the transcriptions using the fantastic free 'keyboard' that my research group and I use
There are also two .csv files in the root directory: dictionary.csv, and survey_data.csv. The former contains meta-information about the word-list sample and its linguistic properties, while the latter has responses to the survey that participants took before recording their pronunciation. Note that many the functions are designed specifically for these datasets.
Find classes & functions in Compare_IPA.py.
The streamlit app was designed for users to explore data (and the relationships between datasets) before jumping into in-depth analysis. The app is still in daily development, however it is in working order.
The study that accompanies the code in this repo was conducted as a final project for Spanish Applied Linguistics: Second Language Acquisition, a course @ The University of Colorado, Denver, taught by my professor & research advisor, Dr. Alyssa Martoccio. Read more about the study here.
git clone https://github.com/djdebonis/IPA_vowel_compare
cd IPA_vowel_compare
pip install .
streamlit run app.py
Dr. Alyssa Martoccio, my professor and research advisor @ CU Denver, taught me most of what I know about the International Phonetic Alphabet (IPA), second language acquisition (SLA), and phonetics in general.
@mathematicalmichael helped me write a lot of data-cleaning functions when I was still early in learning Python. He also introduced me to Streamlit & taught me a lot about Python and computation in general.