A harp with sound waves streaming from its strings; highlighted orange strands represent HARP's prioritized frequency bands

HARP

Harmonic-Aware Residual Partitioning for Neural Audio Codecs

Interspeech 2026 · Oral

Qiaoyu Yang · Lixing He · Binyue Deng · Weifeng Zhao

HARP is a neural audio codec that partitions residual vector quantization across harmonically meaningful frequency bands. A harmonic-aware partitioning distributes codebook capacity over perceptually important bands, so a single model serves multiple bitrates by decoding a growing number of codebook groups. This page presents audio reconstructions across HARP's bitrate tiers.

Audio samples

Each sample shows the reference recording and HARP vs. the DAC baseline at 4.3 and 7.7 kbps.

No samples yet. Generate them with python scripts/build_demo.py your_audio/*.wav, then commit docs/audio/.

Bitrate tiers

GroupsCodebooksBitrate
13~2.6 kbps
25~4.3 kbps
37~6.0 kbps
49~7.7 kbps (full)

Citation

@inproceedings{harp2026,
  title     = {HARP: Harmonic-Aware Residual Partitioning for Neural Audio Codecs},
  author    = {Yang, Qiaoyu and He, Lixing and Deng, Binyue and Zhao, Weifeng},
  booktitle = {Interspeech},
  year      = {2026}
}