Discriminating real and synthetic super-resolved audio samples using embedding-based classifiers

2026-01-06

AI Summary

Although current audio super-resolution models achieve strong performance in perceptual quality and standard metrics, they fail to faithfully reproduce the distributional characteristics of real high-bandwidth audio. This work systematically investigates this limitation by training linear classifiers in multiple audio embedding spaces to distinguish between real and synthetic (generated via GANs or diffusion models) speech and music samples produced by super-resolution systems. Experiments demonstrate that these classifiers consistently achieve near-perfect accuracy across diverse datasets and tasks, revealing a fundamental gap in distributional fidelity among existing methods. This study is the first to quantitatively expose and highlight the critical challenge in audio super-resolution: high perceptual quality does not necessarily imply high distributional fidelity.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Generative adversarial networks (GANs) and diffusion models have recently achieved state-of-the-art performance in audio super-resolution (ADSR), producing perceptually convincing wideband audio from narrowband inputs. However, existing evaluations primarily rely on signal-level or perceptual metrics, leaving open the question of how closely the distributions of synthetic super-resolved and real wideband audio match. Here we address this problem by analyzing the separability of real and super-resolved audio in various embedding spaces. We consider both middle-band ($4\to 16$~kHz) and full-band ($16\to 48$~kHz) upsampling tasks for speech and music, training linear classifiers to distinguish real from synthetic samples based on multiple types of audio embeddings. Comparisons with objective metrics and subjective listening tests reveal that embedding-based classifiers achieve near-perfect separation, even when the generated audio attains high perceptual quality and state-of-the-art metric scores. This behavior is consistent across datasets and models, including recent diffusion-based approaches, highlighting a persistent gap between perceptual quality and true distributional fidelity in ADSR models.

Problem

• audio super-resolution
• distributional fidelity
• synthetic audio detection
• embedding separability
• generative audio models

Innovation

• audio super-resolution
• embedding-based classification
• distributional fidelity
• generative models
• perceptual quality