Perception of Phonological Assimilation by Neural Speech Recognition Models

TL;DR

This paper investigates how neural speech recognition models, specifically Wav2Vec2, perceive and compensate for phonological assimilation, revealing that the model shifts from acoustic to underlying phonological forms influenced by minimal context cues.

Contribution

It provides a systematic analysis of phonological assimilation perception in neural ASR models and uncovers the linguistic knowledge and mechanisms involved.

Findings

The model shifts interpretation from acoustic to underlying phonological form in final layers.

Minimal phonological context cues are sufficient for the model to perform this shift.

Probing experiments reveal the linguistic knowledge implemented in the model.

Abstract

Human listeners effortlessly compensate for phonological changes during speech perception, often unconsciously inferring the intended sounds. For example, listeners infer the underlying /n/ when hearing an utterance such as "clea[m] pan", where [m] arises from place assimilation to the following labial [p]. This article explores how the neural speech recognition model Wav2Vec2 perceives assimilated sounds, and identifies the linguistic knowledge that is implemented by the model to compensate for assimilation during Automatic Speech Recognition (ASR). Using psycholinguistic stimuli, we systematically analyze how various linguistic context cues influence compensation patterns in the model's output. Complementing these behavioral experiments, our probing experiments indicate that the model shifts its interpretation of assimilated sounds from their acoustic form to their underlying form in its final layers. Finally, our causal intervention experiments suggest that the model relies on minimal phonological context cues to accomplish this shift. These findings represent a step towards better understanding the similarities and differences in phonological processing between neural ASR models and humans.