Perception of Phonological Assimilation by Neural Speech Recognition Models
This work addresses the problem of understanding phonological processing in ASR models for researchers in computational linguistics and speech technology, representing an incremental step in comparing model and human perception.
The study investigated how the neural speech recognition model Wav2Vec2 perceives and compensates for phonological assimilation, such as inferring /n/ from [m] in 'clea[m] pan', by analyzing linguistic context cues and finding that the model shifts interpretation from acoustic to underlying forms in its final layers.
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.