Commit ·
a8feebd
1
Parent(s): a71e946
Update README.md
Browse files
README.md
CHANGED
|
@@ -4,7 +4,7 @@ tags:
|
|
| 4 |
license: cc-by-nc-4.0
|
| 5 |
---
|
| 6 |
|
| 7 |
-
# MusicGen - Large -
|
| 8 |
|
| 9 |
Audiocraft provides the code and models for MusicGen, a simple and controllable model for music generation.
|
| 10 |
MusicGen is a single stage auto-regressive Transformer model trained over a 32kHz EnCodec tokenizer with 4 codebooks sampled at 50 Hz.
|
|
@@ -120,10 +120,6 @@ More details on performance measures and human studies can be found in the paper
|
|
| 120 |
|
| 121 |
**Decision thresholds:** Not applicable.
|
| 122 |
|
| 123 |
-
## Data
|
| 124 |
-
|
| 125 |
-
The model is trained on MMI (Meta Music Initiative), Shutterstock, and Pond5.
|
| 126 |
-
|
| 127 |
## Evaluation datasets
|
| 128 |
|
| 129 |
The model was evaluated on the [MusicCaps benchmark](https://www.kaggle.com/datasets/googleai/musiccaps) and on an in-domain held-out evaluation set, with no artist overlap with the training set.
|
|
|
|
| 4 |
license: cc-by-nc-4.0
|
| 5 |
---
|
| 6 |
|
| 7 |
+
# MusicGen - Large - 3.3B
|
| 8 |
|
| 9 |
Audiocraft provides the code and models for MusicGen, a simple and controllable model for music generation.
|
| 10 |
MusicGen is a single stage auto-regressive Transformer model trained over a 32kHz EnCodec tokenizer with 4 codebooks sampled at 50 Hz.
|
|
|
|
| 120 |
|
| 121 |
**Decision thresholds:** Not applicable.
|
| 122 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 123 |
## Evaluation datasets
|
| 124 |
|
| 125 |
The model was evaluated on the [MusicCaps benchmark](https://www.kaggle.com/datasets/googleai/musiccaps) and on an in-domain held-out evaluation set, with no artist overlap with the training set.
|