SILA: Signal-to-Language Augmentation for Enhanced Control in Text-to-Audio Generation

TL;DR

SILA introduces a novel method for fine-grained control over acoustic parameters in text-to-audio generation, enabling more expressive and customizable sound synthesis beyond traditional DSP techniques.

Contribution

The paper presents a model-agnostic approach that learns to disentangle audio semantics from acoustic features, allowing precise control over sound characteristics in generated audio.

Findings

Effective control over loudness, pitch, reverb, and other parameters.

High-quality audio outputs closely match user specifications.

Enhanced versatility for creative sound design.

Abstract

The field of text-to-audio generation has seen significant advancements, and yet the ability to finely control the acoustic characteristics of generated audio remains under-explored. In this paper, we introduce a novel yet simple approach to generate sound effects with control over key acoustic parameters such as loudness, pitch, reverb, fade, brightness, noise and duration, enabling creative applications in sound design and content creation. These parameters extend beyond traditional Digital Signal Processing (DSP) techniques, incorporating learned representations that capture the subtleties of how sound characteristics can be shaped in context, enabling a richer and more nuanced control over the generated audio. Our approach is model-agnostic and is based on learning the disentanglement between audio semantics and its acoustic features. Our approach not only enhances the versatility and expressiveness of text-to-audio generation but also opens new avenues for creative audio production and sound design. Our objective and subjective evaluation results demonstrate the effectiveness of our approach in producing high-quality, customizable audio outputs that align closely with user specifications.