Preprint

Article

What All Do Audio Transformer Models Hear? Probing Acoustic Representations for Language Delivery and Its Structure

Altmetrics

Downloads

420

Views

271

Comments

Yaman Kumar^*,

Jui Shah,Rajiv Ratn Shah,Changyou Chen

Yaman Kumar^*,

Jui Shah,Rajiv Ratn Shah,Changyou Chen

This version is not peer-reviewed

Submitted:

02 January 2021

Posted:

05 January 2021

You are already at the latest version

Alerts

Abstract

In recent times, BERT based transformer models have become an inseparable part of the 'tech stack' of text processing models. Similar progress is being observed in the speech domain with a multitude of models observing state-of-the-art results by using audio transformer models to encode speech. This begs the question of what are these audio transformer models learning. Moreover, although the standard methodology is to choose the last layer embedding for any downstream task, but is it the optimal choice? We try to answer these questions for the two recent audio transformer models, Mockingjay and wave2vec2.0. We compare them on a comprehensive set of language delivery and structure features including audio, fluency and pronunciation features. Additionally, we probe the audio models' understanding of textual surface, syntax, and semantic features and compare them to BERT. We do this over exhaustive settings for native, non-native, synthetic, read and spontaneous speech datasets

Keywords:

Subject: Computer Science and Mathematics - Algebra and Number Theory

Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.