Mind the Prompt: Prompting Strategies in Audio Generations for Improving Sound Classification Article Swipe

PDF

Francesca Ronchini , Ho-Hsiang Wu , Wei-Cheng Lin , Fabio Antonacci ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2504.03329

This paper investigates the design of effective prompt strategies for generating realistic datasets using Text-To-Audio (TTA) models. We also analyze different techniques for efficiently combining these datasets to enhance their utility in sound classification tasks. By evaluating two sound classification datasets with two TTA models, we apply a range of prompt strategies. Our findings reveal that task-specific prompt strategies significantly outperform basic prompt approaches in data generation. Furthermore, merging datasets generated using different TTA models proves to enhance classification performance more effectively than merely increasing the training dataset size. Overall, our results underscore the advantages of these methods as effective data augmentation techniques using synthetic data.

Related Topics

The Coddling Of The American Mind

The Emperor's New Mind

Extended Mind Thesis

Eternal Sunshine Of The Spotless Mind

Society Of Mind

A Beautiful Mind (Film)

Mind

The Game (Mind Game)

Free Your Mind (Song)

Never Mind The Bollocks, Here's The Sex Pistols

Piece Of Mind

Confessions Of A Dangerous Mind (Film)

Concepts

No concepts available.

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2504.03329
PDF: https://arxiv.org/pdf/2504.03329
OA Status: green
OpenAlex ID: https://openalex.org/W4415979406

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4415979406

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2504.03329

Digital Object Identifier
Title: Mind the Prompt: Prompting Strategies in Audio Generations for Improving Sound Classification

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-04-04

Full publication date if available
Authors: Francesca Ronchini, Ho-Hsiang Wu, Wei-Cheng Lin, Fabio Antonacci

List of authors in order
Landing page: https://arxiv.org/abs/2504.03329

Publisher landing page
PDF URL: https://arxiv.org/pdf/2504.03329

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2504.03329

Direct OA link when available
Cited by: 0

Total citation count in OpenAlex

Full payload

id	https://openalex.org/W4415979406
doi	https://doi.org/10.48550/arxiv.2504.03329
ids.doi	https://doi.org/10.48550/arxiv.2504.03329
ids.openalex	https://openalex.org/W4415979406
fwci
type	preprint
title	Mind the Prompt: Prompting Strategies in Audio Generations for Improving Sound Classification
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
is_xpac	False
apc_list
apc_paid
language	en
locations[0].id	pmh:oai:arXiv.org:2504.03329
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2504.03329
locations[0].version	submittedVersion
locations[0].raw_type	text
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2504.03329
locations[1].id	doi:10.48550/arxiv.2504.03329
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2504.03329
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5056089196
authorships[0].author.orcid	https://orcid.org/0000-0001-6897-1645
authorships[0].author.display_name	Francesca Ronchini
authorships[0].author_position	first
authorships[0].raw_author_name	Ronchini, Francesca
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5035643647
authorships[1].author.orcid	https://orcid.org/0000-0002-1102-074X
authorships[1].author.display_name	Ho-Hsiang Wu
authorships[1].author_position	middle
authorships[1].raw_author_name	Wu, Ho-Hsiang
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5070819601
authorships[2].author.orcid	https://orcid.org/0000-0003-1933-1590
authorships[2].author.display_name	Wei-Cheng Lin
authorships[2].author_position	middle
authorships[2].raw_author_name	Lin, Wei-Cheng
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5082156387
authorships[3].author.orcid	https://orcid.org/0000-0003-4545-0315
authorships[3].author.display_name	Fabio Antonacci
authorships[3].author_position	last
authorships[3].raw_author_name	Antonacci, Fabio
authorships[3].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2504.03329
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	Mind the Prompt: Prompting Strategies in Audio Generations for Improving Sound Classification
has_fulltext	False
is_retracted	False
updated_date	2025-11-08T23:21:52.890332
primary_topic
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2504.03329
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2504.03329
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2504.03329
primary_location.id	pmh:oai:arXiv.org:2504.03329
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2504.03329
primary_location.version	submittedVersion
primary_location.raw_type	text
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2504.03329
publication_date	2025-04-04
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	47
abstract_inverted_index.By	35
abstract_inverted_index.We	17
abstract_inverted_index.as	98
abstract_inverted_index.in	31, 64
abstract_inverted_index.of	5, 49, 95
abstract_inverted_index.to	27, 76
abstract_inverted_index.we	45
abstract_inverted_index.Our	52
abstract_inverted_index.TTA	43, 73
abstract_inverted_index.for	9, 22
abstract_inverted_index.our	90
abstract_inverted_index.the	3, 85, 93
abstract_inverted_index.two	37, 42
abstract_inverted_index.This	0
abstract_inverted_index.also	18
abstract_inverted_index.data	65, 100
abstract_inverted_index.more	80
abstract_inverted_index.than	82
abstract_inverted_index.that	55
abstract_inverted_index.with	41
abstract_inverted_index.(TTA)	15
abstract_inverted_index.apply	46
abstract_inverted_index.basic	61
abstract_inverted_index.data.	105
abstract_inverted_index.paper	1
abstract_inverted_index.range	48
abstract_inverted_index.size.	88
abstract_inverted_index.sound	32, 38
abstract_inverted_index.their	29
abstract_inverted_index.these	25, 96
abstract_inverted_index.using	13, 71, 103
abstract_inverted_index.design	4
abstract_inverted_index.merely	83
abstract_inverted_index.models	74
abstract_inverted_index.prompt	7, 50, 57, 62
abstract_inverted_index.proves	75
abstract_inverted_index.reveal	54
abstract_inverted_index.tasks.	34
abstract_inverted_index.analyze	19
abstract_inverted_index.dataset	87
abstract_inverted_index.enhance	28, 77
abstract_inverted_index.merging	68
abstract_inverted_index.methods	97
abstract_inverted_index.models,	44
abstract_inverted_index.models.	16
abstract_inverted_index.results	91
abstract_inverted_index.utility	30
abstract_inverted_index.Overall,	89
abstract_inverted_index.datasets	12, 26, 40, 69
abstract_inverted_index.findings	53
abstract_inverted_index.training	86
abstract_inverted_index.combining	24
abstract_inverted_index.different	20, 72
abstract_inverted_index.effective	6, 99
abstract_inverted_index.generated	70
abstract_inverted_index.realistic	11
abstract_inverted_index.synthetic	104
abstract_inverted_index.advantages	94
abstract_inverted_index.approaches	63
abstract_inverted_index.evaluating	36
abstract_inverted_index.generating	10
abstract_inverted_index.increasing	84
abstract_inverted_index.outperform	60
abstract_inverted_index.strategies	8, 58
abstract_inverted_index.techniques	21, 102
abstract_inverted_index.underscore	92
abstract_inverted_index.effectively	81
abstract_inverted_index.efficiently	23
abstract_inverted_index.generation.	66
abstract_inverted_index.performance	79
abstract_inverted_index.strategies.	51
abstract_inverted_index.Furthermore,	67
abstract_inverted_index.augmentation	101
abstract_inverted_index.investigates	2
abstract_inverted_index.Text-To-Audio	14
abstract_inverted_index.significantly	59
abstract_inverted_index.task-specific	56
abstract_inverted_index.classification	33, 39, 78
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	4
citation_normalized_percentile