This study illustrates how semantic embeddings can be added to and retrieved from NoRaRe. By that, it provides a template for handling vector data and makes popular methodology in semantic modeling available for cross-linguistic comparison.
Author Archives: Arne Rubehn
Generating Phonological Feature Vectors with SoundVectors and CLTS
The recently published Python library soundvectors offers a simple and robust method to derive phonological feature vectors for any valid IPA sound via its canonical description. It is designed to interact neatly with the Cross-Linguistic Transcription Systems reference catalog (CLTS), which dynamically parses valid strings in phonetic transcription to describe speech sounds. This study illustrates how both systems can be used together to generate phonological feature vectors for all kinds of sounds without relying on a previously defined lookup table. Additionally, it compares the generated feature vectors with those obtained from two other prominent databases, PanPhon and PHOIBLE, showing how those systems can be accessed from the CLTS data via its Python API pyclts.