Words Can Shift: Dynamically Adjusting Word Representations Using Nonverbal Behaviors

TL;DR

This paper introduces RAVEN, a model that dynamically adjusts word representations based on nonverbal cues like facial expressions and vocal patterns to better understand human language in face-to-face communication.

Contribution

The paper proposes a novel neural network model that captures fine-grained nonverbal behaviors and shifts word embeddings accordingly, improving multimodal sentiment and emotion analysis.

Findings

Achieves competitive results on sentiment and emotion datasets.

Visualizes how nonverbal cues influence word representations.

Identifies common multimodal patterns in nonverbal contexts.

Abstract

Humans convey their intentions through the usage of both verbal and nonverbal behaviors during face-to-face communication. Speaker intentions often vary dynamically depending on different nonverbal contexts, such as vocal patterns and facial expressions. As a result, when modeling human language, it is essential to not only consider the literal meaning of the words but also the nonverbal contexts in which these words appear. To better model human language, we first model expressive nonverbal representations by analyzing the fine-grained visual and acoustic patterns that occur during word segments. In addition, we seek to capture the dynamic nature of nonverbal intents by shifting word representations based on the accompanying nonverbal behaviors. To this end, we propose the Recurrent Attended Variation Embedding Network (RAVEN) that models the fine-grained structure of nonverbal subword sequences and dynamically shifts word representations based on nonverbal cues. Our proposed model achieves competitive performance on two publicly available datasets for multimodal sentiment analysis and emotion recognition. We also visualize the shifted word representations in different nonverbal contexts and summarize common patterns regarding multimodal variations of word representations.