A computational model of early language acquisition from audiovisual experiences of young infants

TL;DR

This paper introduces a neural network model that learns word segmentation and meanings from real infant-caregiver audiovisual interactions, supporting the idea that infants can bootstrap language acquisition from ambiguous multimodal input.

Contribution

It presents a realistic neural network approach demonstrating how infants might acquire early lexical knowledge from multimodal, referentially ambiguous experiences.

Findings

Model successfully learns word segments and meanings from ambiguous input.

Hidden layers develop phonetic selectivity resembling supervised phone recognition.

Lexical knowledge can emerge from real-world, ambiguous infant-caregiver interactions.

Abstract

Earlier research has suggested that human infants might use statistical dependencies between speech and non-linguistic multimodal input to bootstrap their language learning before they know how to segment words from running speech. However, feasibility of this hypothesis in terms of real-world infant experiences has remained unclear. This paper presents a step towards a more realistic test of the multimodal bootstrapping hypothesis by describing a neural network model that can learn word segments and their meanings from referentially ambiguous acoustic input. The model is tested on recordings of real infant-caregiver interactions using utterance-level labels for concrete visual objects that were attended by the infant when caregiver spoke an utterance containing the name of the object, and using random visual labels for utterances during absence of attention. The results show that beginnings of lexical knowledge may indeed emerge from individually ambiguous learning scenarios. In addition, the hidden layers of the network show gradually increasing selectivity to phonetic categories as a function of layer depth, resembling models trained for phone recognition in a supervised manner.