A Framework for Synthetic Audio Conversations Generation using Large Language Models

TL;DR

ConversaSynth is a framework that leverages large language models and text-to-speech technology to generate diverse, realistic synthetic conversation audio datasets for improving audio AI systems.

Contribution

The paper introduces a novel framework combining LLMs and TTS to produce high-quality synthetic conversational audio datasets for various audio AI tasks.

Findings

Synthetic datasets are diverse and realistic.

Generated audio improves model training and evaluation.

Framework enhances robustness of audio AI systems.

Abstract

In this paper, we introduce ConversaSynth, a framework designed to generate synthetic conversation audio using large language models (LLMs) with multiple persona settings. The framework first creates diverse and coherent text-based dialogues across various topics, which are then converted into audio using text-to-speech (TTS) systems. Our experiments demonstrate that ConversaSynth effectively generates highquality synthetic audio datasets, which can significantly enhance the training and evaluation of models for audio tagging, audio classification, and multi-speaker speech recognition. The results indicate that the synthetic datasets generated by ConversaSynth exhibit substantial diversity and realism, making them suitable for developing robust, adaptable audio-based AI systems.