HILCodec: High-Fidelity and Lightweight Neural Audio Codec
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914955518803968 |
|---|---|
| author | Ahn, Sunghwan Woo, Beom Jun Han, Min Hyun Moon, Chanyeong Kim, Nam Soo |
| author_facet | Ahn, Sunghwan Woo, Beom Jun Han, Min Hyun Moon, Chanyeong Kim, Nam Soo |
| contents | The recent advancement of end-to-end neural audio codecs enables compressing audio at very low bitrates while reconstructing the output audio with high fidelity. Nonetheless, such improvements often come at the cost of increased model complexity. In this paper, we identify and address the problems of existing neural audio codecs. We show that the performance of the SEANet-based codec does not increase consistently as the network depth increases. We analyze the root cause of such a phenomenon and suggest a variance-constrained design. Also, we reveal various distortions in previous waveform domain discriminators and propose a novel distortion-free discriminator. The resulting model, HILCodec, is a real-time streaming audio codec that demonstrates state-of-the-art quality across various bitrates and audio types. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_04752 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | HILCodec: High-Fidelity and Lightweight Neural Audio Codec Ahn, Sunghwan Woo, Beom Jun Han, Min Hyun Moon, Chanyeong Kim, Nam Soo Audio and Speech Processing Sound The recent advancement of end-to-end neural audio codecs enables compressing audio at very low bitrates while reconstructing the output audio with high fidelity. Nonetheless, such improvements often come at the cost of increased model complexity. In this paper, we identify and address the problems of existing neural audio codecs. We show that the performance of the SEANet-based codec does not increase consistently as the network depth increases. We analyze the root cause of such a phenomenon and suggest a variance-constrained design. Also, we reveal various distortions in previous waveform domain discriminators and propose a novel distortion-free discriminator. The resulting model, HILCodec, is a real-time streaming audio codec that demonstrates state-of-the-art quality across various bitrates and audio types. |
| title | HILCodec: High-Fidelity and Lightweight Neural Audio Codec |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2405.04752 |