ADIFF: Explaining audio difference using natural language

Fuente: Zenodo
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Deshmukh, Soham, Han, Shuo, Singh, Rita, Raj, Bhiksha
Format: Recurso digital
Publié: Zenodo 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866902141049765888
author Deshmukh, Soham
Han, Shuo
Singh, Rita
Raj, Bhiksha
author_facet Deshmukh, Soham
Han, Shuo
Singh, Rita
Raj, Bhiksha
contents <div> <div>ADIFF is an audio prefix tuning-based language model with a cross-projection module and undergoes a three-step training process. ADIFF takes two audios and text prompt as input and produces different tiers of difference explanations as output. This involves identifying and describing audio events, acoustic scenes, signal characteristics, and their emotional impact on listeners.</div> <div><br>The code repository is: <a href="https://github.com/soham97/ADIFF">soham97/ADIFF</a></div> </div>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_14706090
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle ADIFF: Explaining audio difference using natural language
Deshmukh, Soham
Han, Shuo
Singh, Rita
Raj, Bhiksha
audio-language
multimodal learning
audio question answering
sound events
acoustic scenes
adiff
zero-shot
<div> <div>ADIFF is an audio prefix tuning-based language model with a cross-projection module and undergoes a three-step training process. ADIFF takes two audios and text prompt as input and produces different tiers of difference explanations as output. This involves identifying and describing audio events, acoustic scenes, signal characteristics, and their emotional impact on listeners.</div> <div><br>The code repository is: <a href="https://github.com/soham97/ADIFF">soham97/ADIFF</a></div> </div>
title ADIFF: Explaining audio difference using natural language
topic audio-language
multimodal learning
audio question answering
sound events
acoustic scenes
adiff
zero-shot
url https://doi.org/10.5281/zenodo.14706090