File size: 1,655 Bytes
868e3b5 bf17a5b 868e3b5 50761e6 868e3b5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 |
# Spirit LM Checkpoints
## Download Checkpoints
Checkpoints are in this repo
Please note that Spirit LM is made available under the **FAIR Noncommercial Research License**
License is here: https://github.com/facebookresearch/spiritlm/blob/main/LICENSE
## Structure
The checkpoints directory should look like this:
```
checkpoints/
βββ README.md
βββ speech_tokenizer
βΒ Β βββ hifigan_spiritlm_base
βΒ Β βΒ Β βββ config.json
βΒ Β βΒ Β βββ generator.pt
βΒ Β βΒ Β βββ speakers.txt
βΒ Β βΒ Β βββ styles.txt
βΒ Β βββ hifigan_spiritlm_expressive_w2v2
βΒ Β βΒ Β βββ config.json
βΒ Β βΒ Β βββ generator.pt
βΒ Β βΒ Β βββ speakers.txt
βΒ Β βββ hubert_25hz
βΒ Β βΒ Β βββ L11_quantizer_500.pt
βΒ Β βΒ Β βββ mhubert_base_25hz.pt
βΒ Β βββ style_encoder_w2v2
βΒ Β βΒ Β βββ config.json
βΒ Β βΒ Β βββ pytorch_model.bin
βΒ Β βββ vqvae_f0_quantizer
βΒ Β βββ config.yaml
βΒ Β βββ model.pt
βββ spiritlm_model
βββ spirit-lm-base-7b
βΒ Β βββ config.json
βΒ Β βββ generation_config.json
βΒ Β βββ pytorch_model.bin
βΒ Β βββ special_tokens_map.json
βΒ Β βββ tokenizer_config.json
βΒ Β βββ tokenizer.model
βββ spirit-lm-expressive-7b
βββ config.json
βββ generation_config.json
βββ pytorch_model.bin
βββ special_tokens_map.json
βββ tokenizer_config.json
βββ tokenizer.model
```
|