Learning Word-Like Units from Joint Audio-Visual Analysis
This enables computers to learn spoken language semantics from audio-visual data, addressing a challenge in multimodal learning for AI systems.
The paper tackles the problem of discovering word-like acoustic units from continuous speech and grounding them to relevant image regions without using speech recognition or text transcriptions, achieving the ability to detect and associate spoken words like 'lighthouse' with corresponding visual objects.
Given a collection of images and spoken audio captions, we present a method for discovering word-like acoustic units in the continuous speech signal and grounding them to semantically relevant image regions. For example, our model is able to detect spoken instances of the word 'lighthouse' within an utterance and associate them with image regions containing lighthouses. We do not use any form of conventional automatic speech recognition, nor do we use any text transcriptions or conventional linguistic annotations. Our model effectively implements a form of spoken language acquisition, in which the computer learns not only to recognize word categories by sound, but also to enrich the words it learns with semantics by grounding them in images.