A Reference-free Metric for Language-Queried Audio Source Separation using Contrastive Language-Audio Pretraining

2024-07-06

arXiv.org

AI Summary

In language-guided audio source separation (LASS), the absence of reference audio and the challenge of jointly evaluating textual semantics and separation quality hinder objective assessment. To address this, we propose CLAPScore—the first reference-free evaluation metric for LASS built upon Contrastive Language–Audio Pretraining (CLAP). CLAPScore computes cross-modal semantic similarity between the separated audio and the input text query, enabling fully reference-free, semantics-driven objective evaluation without requiring ground-truth source signals. Unlike conventional waveform-based metrics such as SDR, CLAPScore is the first to leverage CLAP for LASS evaluation, overcoming the fundamental limitation of text-agnostic metrics. Experiments across multiple LASS benchmarks demonstrate that CLAPScore significantly outperforms SDR and achieves strong agreement with human judgments of semantic relevance (Spearman’s ρ > 0.85). The implementation is publicly available.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Language and Vision

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Large-scale security measurements

Abstract

Language-queried audio source separation (LASS) aims to separate an audio source guided by a text query, with the signal-to-distortion ratio (SDR)-based metrics being commonly used to objectively measure the quality of the separated audio. However, the SDR-based metrics require a reference signal, which is often difficult to obtain in real-world scenarios. In addition, with the SDR-based metrics, the content information of the text query is not considered effectively in LASS. This paper introduces a reference-free evaluation metric using a contrastive language-audio pretraining (CLAP) module, termed CLAPScore, which measures the semantic similarity between the separated audio and the text query. Unlike SDR, the proposed CLAPScore metric evaluates the quality of the separated audio based on the content information of the text query, without needing a reference signal. Experiments show that the CLAPScore provides an effective evaluation of the semantic relevance of the separated audio to the text query, as compared to the SDR metric, offering an alternative for the performance evaluation of LASS systems. The code for evaluation is publicly available.

Problem

• Audio Separation
• Speech Recognition
• Text-audio Enhancement

Innovation

• CLAPScore
• Audio-Text Matching
• LASS Evaluation