arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2502.04770eess.AScs.LG
收藏

神经编解码器中量化效应的高效评估

Efficient Evaluation of Quantization-Effects in Neural Codecs

Cisco Systems, Inc.(思科系统公司)

Wolfgang Mack, Ahmed Mustafa, Rafał Łaganowski, Samer Hijazy

AI总结:

本文提出一种基于模拟数据和低复杂度网络的高效评估框架,用于分析神经编解码器中的量化效应,并据此改进直通估计器的训练稳定性,在内部编解码器和descript-audio-codec上验证了有效性。

摘要:

神经编解码器由编码器、量化器和解码器组成,能够在极低比特率下实现信号传输。训练这些系统需要采用诸如直通估计器、软到硬退火或统计量化器仿真等技术,以使梯度能够非零地跨越量化器传播。评估神经编解码器中量化的影响(例如梯度传递技术对整个系统的影响)通常成本高昂且耗时,原因在于训练需求大以及缺乏经济且可靠的度量指标。本文提出了一种高效的神经编解码器评估框架,该框架使用具有确定比特数的模拟数据和低复杂度神经编码器/解码器来模拟较大网络中的非线性行为。我们的系统在训练时间、计算和硬件需求方面都非常高效,使我们能够揭示神经编解码器中不同的行为特性。基于我们的发现,我们提出了一种改进方法,用于在使用直通估计器时稳定训练过程。我们针对一个内部神经音频编解码器以及最先进的descript-audio-codec验证了我们的发现。

英文摘要:

Neural codecs, comprising an encoder, quantizer, and decoder, enable signal transmission at exceptionally low bitrates. Training these systems requires techniques like the straight-through estimator, soft-to-hard annealing, or statistical quantizer emulation to allow a non-zero gradient across the quantizer. Evaluating the effect of quantization in neural codecs, like the influence of gradient passing techniques on the whole system, is often costly and time-consuming due to training demands and the lack of affordable and reliable metrics. This paper proposes an efficient evaluation framework for neural codecs using simulated data with a defined number of bits and low-complexity neural encoders/decoders to emulate the non-linear behavior in larger networks. Our system is highly efficient in terms of training time and computational and hardware requirements, allowing us to uncover distinct behaviors in neural codecs. We propose a modification to stabilize training with the straight-through estimator based on our findings. We validate our findings against an internal neural audio codec and against the state-of-the-art descript-audio-codec.

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。

↑