arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.12759cs.SDcs.LG
收藏

使用神经音频编解码器的高保真音乐声码器

High-Fidelity Music Vocoder using Neural Audio Codecs

ETH Zurich(苏黎世联邦理工学院)

Luca A. Lanzendörfer, Florian Grötschla, Michael Ungersböck, Roger Wattenhofer

AI总结:

本文提出 DisCoder,通过与 DAC 潜空间对齐的生成对抗编码器-解码器及微调 DAC 解码器,从梅尔频谱图重建高保真 44.1 kHz 音乐,并在客观指标与 MUSHRA 测试中达到最先进水平。

摘要:

尽管神经声码器在高保真语音合成方面取得了显著进展,但其在复调音乐上的应用仍未得到充分探索。在本工作中,我们提出了 DisCoder,这是一种神经声码器,它利用由神经音频编解码器提供信息的生成对抗式编码器-解码器架构,从梅尔频谱图重建高保真 44.1 kHz 音频。我们的方法首先将梅尔频谱图转换为与 Descript Audio Codec(DAC)潜空间对齐的低维表示,然后使用经过微调的 DAC 解码器将其重建为音频信号。DisCoder 在多项客观指标和一项 MUSHRA 听觉研究中取得了音乐合成的最先进性能。我们的方法在语音合成方面也表现出具有竞争力的性能,凸显了其作为通用声码器的潜力。

英文摘要:

While neural vocoders have made significant progress in high-fidelity speech synthesis, their application on polyphonic music has remained underexplored. In this work, we propose DisCoder, a neural vocoder that leverages a generative adversarial encoder-decoder architecture informed by a neural audio codec to reconstruct high-fidelity 44.1 kHz audio from mel spectrograms. Our approach first transforms the mel spectrogram into a lower-dimensional representation aligned with the Descript Audio Codec (DAC) latent space before reconstructing it to an audio signal using a fine-tuned DAC decoder. DisCoder achieves state-of-the-art performance in music synthesis on several objective metrics and in a MUSHRA listening study. Our approach also shows competitive performance in speech synthesis, highlighting its potential as a universal vocoder.

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。

↑