# AudioStory: Generating Long-Form Narrative Audio with Large Language Models

## TL;DR

AudioStory introduces a novel framework combining large language models with text-to-audio systems to generate coherent, long-form narrative audio, addressing the challenge of temporal coherence and compositional reasoning in long audio generation.

## Contribution

It presents a unified, end-to-end trainable system that decomposes complex narratives into sub-tasks and employs a decoupled bridging mechanism for improved coherence and audio quality.

## Key findings

- Outperforms existing TTA baselines in instruction-following and audio fidelity.
- Establishes a new benchmark dataset AudioStory-10K for diverse narrative audio tasks.
- Demonstrates effective long-form audio generation across multiple domains.

## Abstract

Recent advances in text-to-audio (TTA) generation excel at synthesizing short audio clips but struggle with long-form narrative audio, which requires temporal coherence and compositional reasoning. To address this gap, we propose AudioStory, a unified framework that integrates large language models (LLMs) with TTA systems to generate structured, long-form audio narratives. AudioStory possesses strong instruction-following reasoning generation capabilities. It employs LLMs to decompose complex narrative queries into temporally ordered sub-tasks with contextual cues, enabling coherent scene transitions and emotional tone consistency. AudioStory has two appealing features: (1) Decoupled bridging mechanism: AudioStory disentangles LLM-diffuser collaboration into two specialized components, i.e., a bridging query for intra-event semantic alignment and a residual query for cross-event coherence preservation. (2) End-to-end training: By unifying instruction comprehension and audio generation within a single end-to-end framework, AudioStory eliminates the need for modular training pipelines while enhancing synergy between components. Furthermore, we establish a benchmark AudioStory-10K, encompassing diverse domains such as animated soundscapes and natural sound narratives. Extensive experiments show the superiority of AudioStory on both single-audio generation and narrative audio generation, surpassing prior TTA baselines in both instruction-following ability and audio fidelity. Our code is available at https://github.com/TencentARC/AudioStory