AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

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

Yuxin Guo, Teng Wang, Yuying Ge, Shijie Ma, Yixiao Ge, Wei Zou

CVPR 2026 · Large language models › Language, safety and interpretability · approach: Speech and audio language models

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LLM-based instruction decomposition and decoupled bridging (AudioStory) versus LLM+TangoFlux captioningTangoFlux · AS-10k long-form narrative audioCLAP: 0.322 -> 0.392Improved
add storyline reasoning before generation (ablation)AudioStory · AS-10kinstruction-following score: 3.1 -> 4.1Improved
add explicit per-clip captions (interleaved reasoning generation) (ablation)AudioStory · AS-10kFAD: 16.03 -> 3.06Improved
supervise semantic tokens with T5 text instead of AudioMAE/Whisper audio features (ablation)AudioStory bridging · Single audio generationbridging ablation score: 9.55 -> 2.29 (AudioMAE vs T5)Improved
supervise residual tokens weakly via DiT flow loss instead of explicit audio-feature supervision (ablation)AudioStory bridging · Single audio generationbridging ablation score: 3.71 -> 2.29 (Whisper-guided vs ours)Improved
fully unfreeze DiT in joint training instead of freezing it (ablation)AudioStory · Single audio generationFAD: 11.04 -> 2.29Improved
unfreeze all DiT rather than only Single-DiT or only MM-DiT (ablation)AudioStory · Single audio generationFAD: 5.84 / 3.21 -> 2.29Improved
add residual tokens to the bridge in joint training (ablation)AudioStory · Single audio generationFAD: 4.66 -> 2.29Improved
train generation first then understanding (stage order G then U) instead of understanding first (ablation)AudioStory · Long-form audio generationFAD: 9.53 -> 3.00Improved
add DiT joint training in Stage-II before understanding stage versus after (ablation)AudioStory · Long-form audio generationFAD: 6.50 -> 3.00Improved
use progressive training instead of joint training in one stage (ablation)AudioStory · Long-form audio generationFAD: 8.74 -> 3.00Improved
AudioStory-Base versus TangoFlux on single-audio CLAPTangoFlux · AudioCaps testCLAP: 0.385 -> 0.383No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy