# A sound description: Exploring prompt templates and class descriptions to enhance zero-shot audio classification

arXiv: https://arxiv.org/abs/2409.13676 · HTML: https://mlsanity.com/paper/2409.13676/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 28 references that are in the map.
- First posted on arXiv 19 Sep 2024.

## Builds on (most cited first, 22 of 22)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [AudioLDM: Text-to-Audio Generation with Latent Diffusion Models](https://mlsanity.com/paper/2301.12503/) (2023)
- [FSD50K: An Open Dataset of Human-Labeled Sound Events](https://mlsanity.com/paper/2010.00475/) (2020)
- [AudioCLIP: Extending CLIP to Image, Text and Audio](https://mlsanity.com/paper/2106.13043/) (2021)
- [What does a platypus look like? Generating customized prompts for zero-shot image classification](https://mlsanity.com/paper/2209.03320/) (2022)
- [WavCaps: A ChatGPT-Assisted Weakly-Labelled Audio Captioning Dataset for Audio-Language Multimodal Research](https://mlsanity.com/paper/2303.17395/) (2023)
- [HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection](https://mlsanity.com/paper/2202.00874/) (2022)
- [Waffling around for Performance: Visual Classification with Random Words and Broad Concepts](https://mlsanity.com/paper/2306.07282/) (2023)
- [The Butterfly Effect of Altering Prompts: How Small Changes and Jailbreaks Affect Large Language Model Performance](https://mlsanity.com/paper/2401.03729/) (2024)
- [Natural Language Supervision for General-Purpose Audio Representations](https://mlsanity.com/paper/2309.05767/) (2023)
- [LaFTer: Label-Free Tuning of Zero-shot Classifier using Language and Unlabeled Image Collections](https://mlsanity.com/paper/2305.18287/) (2023)
- [Contrastive Audio-Language Learning for Music](https://mlsanity.com/paper/2208.12208/) (2022)
- [PerceptionCLIP: Visual Classification by Inferring and Conditioning on Contexts](https://mlsanity.com/paper/2308.01313/) (2023)
- [Audio-free Prompt Tuning for Language-Audio Models](https://mlsanity.com/paper/2309.08357/) (2023)
- [Domain Adaptation for Contrastive Audio-Language Models](https://mlsanity.com/paper/2402.09585/) (2024)
- [A Multimodal Prototypical Approach for Unsupervised Sound Classification](https://mlsanity.com/paper/2306.12300/) (2023)
- [A SOUND APPROACH: Using Large Language Models to generate audio descriptions for egocentric text-audio retrieval](https://mlsanity.com/paper/2402.19106/) (2024)

## Built on by (most cited first, 3 of 3)

- [Leveraging Prediction Entropy for Automatic Prompt Weighting in Zero-Shot Audio-Language Classification](https://mlsanity.com/paper/2601.05011/) (2026)
- [CALM: Class-Conditional Sparse Attention Vectors for Large Audio-Language Models](https://mlsanity.com/paper/2602.07077/) (2026)
- [Prompt Amplification and Zero-Shot Late Fusion in Audio-Language Models for Speech Emotion Recognition](https://mlsanity.com/paper/2603.23057/) (2026)

## Abstract (arXiv)

Audio-text models trained via contrastive learning offer a practical approach to perform audio classification through natural language prompts, such as "this is a sound of" followed by category names. In this work, we explore alternative prompt templates for zero-shot audio classification, demonstrating the existence of higher-performing options. First, we find that the formatting of the prompts significantly affects performance so that simply prompting the models with properly formatted class labels performs competitively with optimized prompt templates and even prompt ensembling. Moreover, we look into complementing class labels by audio-centric descriptions. By leveraging large language models, we generate textual descriptions that prioritize acoustic features of sound events to disambiguate between classes, without extensive prompt engineering. We show that prompting with class descriptions leads to state-of-the-art results in zero-shot audio classification across major ambient sound datasets. Remarkably, this method requires no additional training and remains fully zero-shot.