Skip to content
AI News HubLIVE
Original source2 min read

From Visual Cues to Spoken Narration: Rethinking Audio Description

Summary

The paper introduces Cue2Narrate, a two-stage pipeline that jointly predicts what to narrate and when to insert audio description in long untrimmed movie clips. It also presents the LongLSMDC benchmark with clips averaging about 6.5 minutes, and shows gains of 5–12 points in average mAP over video-only and audio-only localization baselines.

SourcearXiv Computer VisionAuthor: Akshita Gupta, Aditya Arora, Federico Tombari, Marcus Rohrbach, Anna Rohrbach
From Visual Cues to Spoken Narration: Rethinking Audio Description
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 1 Sep 2026]

Title:From Visual Cues to Spoken Narration: Rethinking Audio Description

View a PDF of the paper titled From Visual Cues to Spoken Narration: Rethinking Audio Description, by Akshita Gupta and 4 other authors

View PDF HTML (experimental)

Abstract:Audio Description (AD) provides spoken narration of visual events during dialogue gaps, making movies accessible to visually impaired audiences. The problem requires determining both what (which visual event) and when (position for inserting the AD) to narrate, to achieve the best user experience. Prior work has largely reduced the problem to video captioning of pre-segmented video clips, i.e., what is largely predefined and when is ignored entirely. We propose Cue2Narrate, a two-stage pipeline that jointly predicts what and when to narrate in longer untrimmed movie clips. A dual-head audio-visual localizer predicts two temporally distinct windows per AD utterance: a visual cue window and a spoken narration window. A LoRA-adapted VLM then generates concise ADs from the predicted visual evidence, trained with a Description Ranking Loss that ranks captions (negative samples) of the same frames lower than the GT AD. To benchmark this new problem statement, we introduce the LongLSMDC benchmark with up to 8-min movie clips (~6.5min on average). On LongLSMDC, Cue2Narrate outperforms video-only and audio-only localization baselines by 5--12 points in avg. mAP. Under both predicted- and GT-window evaluation, Cue2Narrate improves AD generation over the corresponding fine-tuned base VLM. These results establish the first benchmark for multi-segment AD generation on long-form clips. Data & Code: this https URL

Comments: Accepted to EMNLP main conference

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2609.01725 [cs.CV]

(or arXiv:2609.01725v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2609.01725

arXiv-issued DOI via DataCite

Submission history

From: Akshita Gupta [view email] [v1] Tue, 1 Sep 2026 18:00:50 UTC (13,448 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled From Visual Cues to Spoken Narration: Rethinking Audio Description, by Akshita Gupta and 4 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-09

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • Cue2Narrate uses a dual-head audio-visual localizer to predict a visual cue window and a spoken narration window for each AD utterance.
  • A LoRA-adapted VLM generates concise ADs from predicted visual evidence, trained with a Description Ranking Loss.
  • LongLSMDC is the first benchmark for multi-segment AD generation on long-form clips, with clips up to 8 minutes and ~6.5 minutes on average.
  • The method outperforms video-only and audio-only localization baselines by 5–12 points in average mAP.

Highlights and analysis are generated automatically and may contain errors. Check the original source.