AudioLCM

Text-to-Audio Generation with Latent Consistency Models

paper|code|

Huadai Liu, Rongjie Huang, Yang Liu, Hengyuan Cao, Jialei Wang, Xize Cheng, Siqi Zheng, Zhou Zhao

Abstract. Recent advancements in Latent Diffusion Models (LDMs) have propelled them to the forefront of various generative tasks. However, their iterative sampling process poses a significant computational burden, resulting in slow generation speeds and limiting their application in text-to-audio generation deployment. In this work, we introduce AudioLCM, a novel consistency-based model tailored for efficient and high-quality text-to-audio generation. Unlike prior approaches that address noise removal through iterative processes, AudioLCM integrates Consistency Models (CMs) into the generation process, facilitating rapid inference through a mapping from any point at any time step to the trajectory's initial point. To overcome the convergence issue inherent in LDMs with reduced sample iterations, we propose the Guided Latent Consistency Distillation with a multi-step Ordinary Differential Equation (ODE) solver. This innovation shortens the time schedule from thousands to dozens of steps while maintaining sample quality, thereby achieving fast convergence and high-quality generation. Furthermore, to optimize the performance of transformer-based neural network architectures, we integrate the advanced techniques pioneered by LLaMA into the foundational framework of transformers. This architecture supports stable and efficient training, ensuring robust performance in text-to-audio synthesis. Experimental results on text-to-audio generation and text-to-music synthesis tasks demonstrate that AudioLCM needs only 2 iterations to synthesize high-fidelity audios, while it maintains sample quality competitive with state-of-the-art models using hundreds of steps. AudioLCM enables a sampling speed of 333x faster than real-time on a single NVIDIA 4090Ti GPU, making generative models practically applicable to text-to-audio generation deployment. Our extensive preliminary analysis shows that each design in AudioLCM is effective.

AudioLCM Overview

An illustration of EchoSpeech. AudioLCM propose the Guided Consistency Distillation with k-step ODE solver. c is the text embedding and 𝜔 is the classifier-free guidance scale.

Table of Contents

Text-to-Audio Generation

                       Text Prompts                        Ground-truth AudioLCM Teacher Make-an-audio2 AudioLDM2 Tango AudioLDM Make-an-Audio AudioGen

Text-to-Music Generation

                       Text Prompts                        Ground-truth AudioLCM Teacher AudioLDM2 MusicLDM MusicGen Riffusion

Preliminary Analyses - Classifier-free Guidance

                       Text Prompts                        Ground-truth Scale 1 Scale 3 Scale 5 Scale 7 Scale 9

Preliminary Analyses - Multi-step ODE Solver

                       Text Prompts                        Ground-truth k 1 k 5 k 10 k 20 k 50