We lifted 12 functions out of this paper's own repositories and ran 10 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| aask1357/hilcodec | canonical | 10 of 12 |
| Function | Status | Where it lives |
|---|---|---|
| SConv1d | Ran | aask1357/hilcodec/models/hilcodec/causal_layers.py code served (permissive licence) · get_code("7127cd68fc0b6181") |
| SConvTranspose1d | Ran | aask1357/hilcodec/models/hilcodec/causal_layers.py code served (permissive licence) · get_code("2c73c22b72590d06") |
| get_2d_padding | Ran | aask1357/hilcodec/models/hilcodec/discriminators.py code served (permissive licence) · get_code("1473010bb86abf65") |
| get_padding | Ran | aask1357/hilcodec/models/hilcodec/avocodo.py code served (permissive licence) · get_code("e28f4ff393bc05c9") |
| kmeans | Ran | aask1357/hilcodec/models/hilcodec/vector_quantize.py code served (permissive licence) · get_code("5a866a6652d3e867") |
| l2norm | Ran | aask1357/hilcodec/models/hilcodec/vector_quantize.py code served (permissive licence) · get_code("0254d06d15bac873") |
| log_clip | Ran | aask1357/hilcodec/models/hilcodec/losses.py code served (permissive licence) · get_code("0a1977d56c184638") |
| mel_scale | Ran | aask1357/hilcodec/models/hilcodec/losses.py code served (permissive licence) · get_code("da36a93a8ef96944") |
| mel_spectrogram | Ran | aask1357/hilcodec/models/hilcodec/losses.py code served (permissive licence) · get_code("221f45f54c4ff8f5") |
| sample_vectors | Ran | aask1357/hilcodec/models/hilcodec/vector_quantize.py code served (permissive licence) · get_code("01c3ffb2465073f7") |
| get_extra_padding_for_conv1d | Not yet run | aask1357/hilcodec/models/hilcodec/modules/conv.py code served (permissive licence) · get_code("8d9aa381c6147164") |
| get_norm_module | Not yet run | aask1357/hilcodec/models/hilcodec/modules/conv.py code served (permissive licence) · get_code("5871424a686367d5") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
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.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2405.04752")
get_code_for_paper("2405.04752")
have("2405.04752")
Connect an agent — have() is free.