Published January 29, 2023
| Version v4
Preprint
Open
AudioLDM: Text-to-Audio Generation with Latent Diffusion Models
Creators
- 1. University of Surrey
- 2. Imperial College London
Description
This space host the pre-trained model of AudioLDM
audioldm-m-text-ft (**recommand**, default): the medium large AudioLDM finetuned with AudioCaps and MusicCaps audio-text pairs *(added 2023-04-10)*.
audioldm-s-text-ft (**recommand**): the small AudioLDM finetuned with AudioCaps and MusicCaps audio-text pairs *(added 2023-04-10)*.
audioldm-m-full: the medium AudioLDM without finetuning and trained with audio embeddings as condition *(added 2023-04-10)*.
audioldm-s-full-v2: more training steps comparing with audioldm-s-full *(added 2023-03-04)*.
audioldm-l-full: larger model comparing with audioldm-s-full *(added 2023-03-04)*.
audioldm-s-full: the original open-sourced version *(added 2023-02-01)*.
Files
hifigan-vocoder-config.json
Files
(25.0 GB)
| Name | Size | Download all |
|---|---|---|
|
md5:8c62081daf5a1b2b9fa014e5a56ff03f
|
7.0 GB | Download |
|
md5:fca6e5a37c7bf47a701a662d48aeded8
|
2.6 GB | Download |
|
md5:46bad9f176651404b3cf1484942749b9
|
4.6 GB | Download |
|
md5:036bc9b547a50f78b960ef8f14d0e1fb
|
4.6 GB | Download |
|
md5:5451d3628b3a2b535f87044685fe102c
|
2.6 GB | Download |
|
md5:70a662bf2d200cdd1913c5533b126007
|
2.6 GB | Download |
|
md5:8d2ad944f74cf71cdf140dac9f46728c
|
767 Bytes | Preview Download |
|
md5:77a0f15a947821b999dbc0e22fea0a33
|
221.2 MB | Download |
|
md5:77fcff0cdf9073dc8cd1bd1db268b228
|
978.0 MB | Download |