CVAug 21, 2023

Can Language Models Learn to Listen?

Berkeley
arXiv:2308.10897v146 citationsh-index: 85
Originality Incremental advance
AI Analysis

This work addresses the challenge of creating realistic listener behaviors in human-computer interaction, though it is incremental as it adapts existing methods to a new domain.

The paper tackles the problem of generating appropriate facial responses for a listener in dyadic social interactions based on speaker transcriptions, achieving significantly higher quality responses by initializing a transformer with pre-trained language model weights.

We present a framework for generating appropriate facial responses from a listener in dyadic social interactions based on the speaker's words. Given an input transcription of the speaker's words with their timestamps, our approach autoregressively predicts a response of a listener: a sequence of listener facial gestures, quantized using a VQ-VAE. Since gesture is a language component, we propose treating the quantized atomic motion elements as additional language token inputs to a transformer-based large language model. Initializing our transformer with the weights of a language model pre-trained only on text results in significantly higher quality listener responses than training a transformer from scratch. We show that our generated listener motion is fluent and reflective of language semantics through quantitative metrics and a qualitative user study. In our evaluation, we analyze the model's ability to utilize temporal and semantic aspects of spoken text. Project page: https://people.eecs.berkeley.edu/~evonne_ng/projects/text2listen/

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes