探索基于Transformer的文本到音乐(TTM)模型的压缩性
Exploring compressibility of transformer based text-to-music (TTM) models
AI总结:
本文研究基于Transformer的文本到音乐模型在手机部署面临的算力问题,通过知识蒸馏和组件修改进行压缩,提出TinyTTM模型,在显著减少参数的同时取得了优于MusicGen-Small的生成性能。
摘要:
最先进的Text-To-Music(文本到音乐,TTM)生成式AI模型体积庞大,需要桌面或服务器级算力,使其无法在手机上部署。本文分析了TTM模型在模型压缩与生成性能之间的权衡。我们通过知识蒸馏及特定修改来研究压缩,使其适用于TTM模型的各个组件(编码器、生成模型和解码器)。利用这些方法,我们创建了TinyTTM(89.2M参数),在MusicBench数据集上达到3.66的FAD和1.32的KL,优于MusicGen-Small(557.6M参数),但未低于在MusicBench上微调的MusicGen-small。
英文摘要:
State-of-the art Text-To-Music (TTM) generative AI models are large and require desktop or server class compute, making them infeasible for deployment on mobile phones. This paper presents an analysis of trade-offs between model compression and generation performance of TTM models. We study compression through knowledge distillation and specific modifications that enable applicability over the various components of the TTM model (encoder, generative model and the decoder). Leveraging these methods we create TinyTTM (89.2M params) that achieves a FAD of 3.66 and KL of 1.32 on MusicBench dataset, better than MusicGen-Small (557.6M params) but not lower than MusicGen-small fine-tuned on MusicBench.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。