We lifted 16 functions out of this paper's own repositories and ran 8 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 |
|---|---|---|
| mathieulagrange/dcasefadembedding | canonical | 2 of 2 |
| YoonjinXD/kadtk | pwc_unofficial | 4 of 12 |
| dcase2024-task7-sound-scene-synthesis/fadtk | extension | 2 of 2 |
| Function | Status | Where it lives |
|---|---|---|
| calc_embd_statistics | Ran | dcase2024-task7-sound-scene-synthesis/fadtk/fadtk/fad.py code served (permissive licence) · get_code("90c857014df5f615") |
| create_prompt | Ran | dcase2024-task7-sound-scene-synthesis/fadtk/example/prompts/gpt4_quality.py code served (permissive licence) · get_code("ca4b05e19c89061a") |
| custom_float_format | Ran | mathieulagrange/dcasefadembedding/correlation_calculation.py code served (permissive licence) · get_code("9dc1158acf7f87d1") |
| do_mixup | Ran | YoonjinXD/kadtk/kadtk/models/panns/pytorch_utils.py code served (permissive licence) · get_code("37ad96806353e8cb") |
| get_cache_embedding_path | Ran | YoonjinXD/kadtk/kadtk/utils.py code served (permissive licence) · get_code("c8e9bc934b6ab42c") |
| get_noisy_correlation_table | Ran | mathieulagrange/dcasefadembedding/correlation_calculation.py code served (permissive licence) · get_code("c7adc236598ac778") |
| median_pairwise_distance | Ran | YoonjinXD/kadtk/kadtk/kad.py code served (permissive licence) · get_code("2bb4926b81a065da") |
| move_data_to_device | Ran | YoonjinXD/kadtk/kadtk/models/panns/pytorch_utils.py code served (permissive licence) · get_code("43a8c72263d56894") |
| calc_frechet_distance | Not yet run | YoonjinXD/kadtk/kadtk/fad.py code served (permissive licence) · get_code("e9e2298075d48eb9") |
| calc_kernel_audio_distance | Not yet run | YoonjinXD/kadtk/kadtk/kad.py code served (permissive licence) · get_code("1043d637cef813a6") |
| forward | Not yet run | YoonjinXD/kadtk/kadtk/models/panns/pytorch_utils.py code served (permissive licence) · get_code("20581601538964cf") |
| get_cache_embedding_paths | Not yet run | YoonjinXD/kadtk/kadtk/utils.py code served (permissive licence) · get_code("f50a75ff82ad0dcd") |
| get_image_embedding | Not yet run | YoonjinXD/kadtk/kadtk/models/openl3/core.py code served (permissive licence) · get_code("9829300e384d092d") |
| kapre_v0_1_4_magnitude_to_decibel | Not yet run | YoonjinXD/kadtk/kadtk/models/openl3/models.py code served (permissive licence) · get_code("a1560674b26e5a3b") |
| load_audio_embedding_model | Not yet run | YoonjinXD/kadtk/kadtk/models/openl3/models.py code served (permissive licence) · get_code("e86a92fe6747fef9") |
| load_audio_embedding_model_from_path | Not yet run | YoonjinXD/kadtk/kadtk/models/openl3/models.py code served (permissive licence) · get_code("a1ab6fe4517a082d") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
This paper explores whether considering alternative domain-specific embeddings to calculate the Fréchet Audio Distance (FAD) metric can help the FAD to correlate better with perceptual ratings of environmental sounds. We used embeddings from VGGish, PANNs, MS-CLAP, L-CLAP, and MERT, which are tailored for either music or environmental sound evaluation. The FAD scores were calculated for sounds from the DCASE 2023 Task 7 dataset. Using perceptual data from the same task, we find that PANNs-WGM-LogMel produces the best correlation between FAD scores and perceptual ratings of both audio quality and perceived fit with a Spearman correlation higher than 0.5. We also find that music-specific embeddings resulted in significantly lower results. Interestingly, VGGish, the embedding used for the original Fréchet calculation, yielded a correlation below 0.1. These results underscore the critical importance of the choice of embedding for the FAD metric design.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2403.17508")
get_code_for_paper("2403.17508")
have("2403.17508")
Connect an agent — have() is free.