Human Speech Perception in Noise: Can Large Language Models Paraphrase to Improve It?

TL;DR

This paper explores whether large language models can generate paraphrases that improve human speech perception in noisy environments, proposing a new prompting method that significantly enhances intelligibility.

Contribution

It introduces a novel task of generating acoustically intelligible paraphrases using LLMs and proposes a prompt-and-select approach to decouple textual and non-textual attributes.

Findings

Prompt-and-select improves speech perception by 40% in noisy conditions.

LLMs struggle to control acoustic intelligibility with standard prompts.

The method enhances human perception of distorted speech in noise.

Abstract

Large Language Models (LLMs) can generate text by transferring style attributes like formality resulting in formal or informal text. However, instructing LLMs to generate text that when spoken, is more intelligible in an acoustically difficult environment, is an under-explored topic. We conduct the first study to evaluate LLMs on a novel task of generating acoustically intelligible paraphrases for better human speech perception in noise. Our experiments in English demonstrated that with standard prompting, LLMs struggle to control the non-textual attribute, i.e., acoustic intelligibility, while efficiently capturing the desired textual attributes like semantic equivalence. To remedy this issue, we propose a simple prompting approach, prompt-and-select, which generates paraphrases by decoupling the desired textual and non-textual attributes in the text generation pipeline. Our approach resulted in a 40% relative improvement in human speech perception, by paraphrasing utterances that are highly distorted in a listening condition with babble noise at a signal-to-noise ratio (SNR) -5 dB. This study reveals the limitation of LLMs in capturing non-textual attributes, and our proposed method showcases the potential of using LLMs for better human speech perception in noise.