# Massively Multilingual Adversarial Speech Recognition

## TL;DR

This paper explores training multilingual speech recognition models on up to 100 languages, highlighting the impact of linguistic similarities and introducing adversarial objectives to improve language independence.

## Contribution

It introduces novel pretraining objectives that enhance language-independent representations in large-scale multilingual speech recognition models.

## Key findings

- Pretraining on multiple languages improves recognition accuracy.
- Language similarity influences model performance.
- Adversarial objectives promote language-independent encoder features.

## Abstract

We report on adaptation of multilingual end-to-end speech recognition models trained on as many as 100 languages. Our findings shed light on the relative importance of similarity between the target and pretraining languages along the dimensions of phonetics, phonology, language family, geographical location, and orthography. In this context, experiments demonstrate the effectiveness of two additional pretraining objectives in encouraging language-independent encoder representations: a context-independent phoneme objective paired with a language-adversarial classification objective.