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 2602.15491cs.SDcs.AI
收藏

均衡器:在神经音频编解码器中引入形状-增益分解

The Equalizer: Introducing Shape-Gain Decomposition in Neural Audio Codecs

Inria at Univ. Grenoble Alpes, CNRS, LJK, France(法国 Grenoble 阿尔卑斯大学 Inria 分校、CNRS、LJK);Univ. Grenoble Alpes, Grenoble-INP, GIPSA-lab, France(法国 Grenoble 阿尔卑斯大学、Grenoble-INP、GIPSA-lab)

Samir Sadok, Laurent Girin, Xavier Alameda-Pineda

AI总结:

本研究提出在神经音频编解码器中引入形状-增益分解,以提高编码效率和比特率-失真性能。

摘要:

神经音频编解码器(NACs)通常将语音/音频信号的短时能量(增益)和归一化结构(形状)在同一个潜在空间中联合编码。因此,它们在输入信号水平的全局变化方面表现较差,因为这种变化对编码器输出的嵌入向量及其量化有很强的影响。这种技术本质上效率低下,导致代码本冗余和次优的比特率-失真性能。为了解决这些限制,我们提出将广泛用于经典语音/音频编码中的形状-增益分解引入NAC框架中。所提出均衡器方法的原则是在NAC编码器之前,将输入信号按短时基线分解为增益和归一化形状向量。形状向量由NAC处理,而增益则通过标量量化并单独传输。输出(解码)信号从NAC的归一化输出和量化增益重新构建。我们在语音信号上的实验表明,这种通用方法,易于应用于任何NAC,能够显著提高比特率-失真性能,并大幅减少复杂性。

英文摘要:

Neural audio codecs (NACs) typically encode the short-term energy (gain) and normalized structure (shape) of speech/audio signals jointly within the same latent space. As a result, they are poorly robust to a global variation of the input signal level in the sense that such a variation has a strong influence on the embedding vectors at the output of the encoder and their quantization. This methodology is inherently inefficient, leading to codebook redundancy and suboptimal bitrate-distortion performance. To address these limitations, we propose to introduce shape-gain decomposition, widely used in classical speech and audio coding, into the NAC framework. The principle of the proposed Equalizer methodology, easily applicable to any NAC, is to decompose the input signal---before the NAC encoder---into gain and normalized shape vector on a short-term basis. The shape vector is processed by the NAC, while the gain is quantized with scalar quantization and transmitted separately. The output (decoded) signal is reconstructed from the normalized output of the NAC and the quantized gain. Our experiments conducted on speech signals with four different prominent codecs show that this general methodology enables a substantial gain in bitrate-distortion performance, as well as a massive reduction in quantizer complexity.

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。

↑