提出新型字节级分词法，实现24位高保真音频的可训练无损压缩。
Benchmarking Language Modeling for Lossless Compression of Full-Fidelity Audio
- 采用字节级分词，解决高比特率下词汇表爆炸问题
- 首次实现24位音频的可训练无损压缩，性能优于FLAC
- 适用于音乐、语音等多领域高保真音频压缩
基于原始波形训练的自回归“语言”模型可用于无损音频压缩，但以往研究仅限于8位音频，未验证其在16/24位实际场景下的可行性及与现有编解码器的竞争力。本文在多样化的音频领域（音乐、语音、生物声学）、采样率（16kHz-48kHz）和比特深度（8、16、24位）上对基于语言模型的压缩方法进行基准测试。标准样本级分词在高比特深度下因词汇表规模过大（16位为65K，24位达1670万）而不可行。为此，我们提出Trilobyte，一种面向全分辨率音频的字节级分词方案，将词汇表扩展从$O(2^{b})$降至$O(1)$，首次实现24位语言模型驱动的无损压缩。实验表明，尽管语言模型在8位和16位下持续优于FLAC并达到当前最优，但在超过8位后压缩增益趋于平缓。
原文摘要 · Abstract (English)
Autoregressive "language" models (LMs) trained on raw waveforms can be repurposed for lossless audio compression, but prior work is limited to 8-bit audio, leaving open whether such approaches work for practical settings (16/24-bit) and can compete with existing codecs. We benchmark LM-based compression on full-fidelity audio across diverse domains (music, speech, bioacoustics), sampling rates (16kHz-48kHz), and bit depths (8, 16, 24-bit). Standard sample-level tokenization becomes intractable at higher bit depths due to vocabulary size (65K for 16-bit; 16.7M for 24-bit). We propose Trilobyte, a byte-level tokenization schema for full resolution audio, improving vocabulary scaling from $O(2^{b})$ to $O(1)$ and enabling the first tractable 24-bit LM-based lossless compression. While LMs consistently outperform FLAC and yield state-of-the-art compression at 8-bit and 16-bit, we observe that compression gains become more modest as bit depth increases beyond 8-bit.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。