Search is not available for this dataset
audio audio |
|---|
End of preview. Expand in Data Studio
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
TrueMuse — Sample Dataset
This is a representative sample of the TrueMuse benchmark, provided for reviewer inspection. It covers all concept types (genre, melody, musician, instrument) and all three generative models (AudioLDM2, Mustango, Stable Audio), with 1 audio file per sub-folder to illustrate the dataset structure. Metadata files and embeddings are not included.
Full Dataset
The complete dataset is available at: https://huggingface.co/datasets/AnonymousAuthorsssss/TrueMuse_A_Benchmark_for_Data_Attribution_in_Text_to_Music_Models
Sample Structure
data/
├── concepts/ # Reference audio clips per concept instance
│ ├── genre_3/ # 8 genre instances (3-clip training set)
│ │ ├── electronic/
│ │ ├── folk/
│ │ └── ...
│ ├── genre_6/ # 8 genre instances (6-clip training set)
│ ├── melody_1/ # 50 melody instances
│ │ ├── melody_001/
│ │ └── ...
│ ├── musician_3/ # 50 musician instances (3-clip)
│ └── musician_6/ # 50 musician instances (6-clip)
└── generated/ # Model-generated audio
├── audioldm2/
│ ├── genre_3/
│ │ └── electronic_ldm2_db/
│ │ ├── contextual/ # 1 sample audio
│ │ └── style/ # 1 sample audio
│ ├── instrument_3/ # 25 instrument instances
│ ├── instrument_6/ # 25 instrument instances
│ ├── melody_1/
│ ├── musician_3/
│ └── musician_6/
├── mustango/ # same structure as audioldm2
└── stable-audio/ # same structure as audioldm2
- Downloads last month
- 8