# Probabilistic Typology: Deep Generative Models of Vowel Inventories

## TL;DR

This paper introduces deep probabilistic models to analyze vowel inventories in languages, aiming to identify universal patterns and variations in phonological typology through extensive experiments on over 200 languages.

## Contribution

It presents the first probabilistic approach to phonological typology using deep stochastic point processes, advancing beyond previous simulation-based methods.

## Key findings

- Deep generative models effectively capture vowel inventory patterns.
- The models reveal universal tendencies and language-specific variations.
- Experimental results on 200+ languages validate the approach.

## Abstract

Linguistic typology studies the range of structures present in human language. The main goal of the field is to discover which sets of possible phenomena are universal, and which are merely frequent. For example, all languages have vowels, while most---but not all---languages have an /u/ sound. In this paper we present the first probabilistic treatment of a basic question in phonological typology: What makes a natural vowel inventory? We introduce a series of deep stochastic point processes, and contrast them with previous computational, simulation-based approaches. We provide a comprehensive suite of experiments on over 200 distinct languages.