Instructions to use ScreamingTony/AffectF5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- F5-TTS
How to use ScreamingTony/AffectF5 with F5-TTS:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Download config.yaml from ScreamingTony/AffectF5: direct link, hf CLI and curl.
- Browser
- Download file 1.59 kB
-
https://huggingface.co/ScreamingTony/AffectF5/resolve/main/config.yaml
- Command line
-
hf download hf://ScreamingTony/AffectF5/config.yaml
-
curl -L -o config.yaml https://huggingface.co/ScreamingTony/AffectF5/resolve/main/config.yaml
1.59 kB
| base: | |
| repo_id: SWivid/F5-TTS | |
| subfolder: F5TTS_v1_Base | |
| checkpoint: model_1250000.safetensors | |
| sha256: 670900fd14e6c458b95da6e9ed317cdb20dbaf7a1c02ac06a05475a9d32b6a38 | |
| vocab: vocab.txt | |
| adapter: | |
| repo_id: ScreamingTony/AffectF5 | |
| filename: affectf5_c7_adapter.safetensors | |
| sha256: 26dd9fdc69395e8a937feadb3417e5503667eaedb6a85b12b26d0374a58f253f | |
| model: | |
| dim: 1024 | |
| depth: 22 | |
| heads: 16 | |
| ff_mult: 2 | |
| text_dim: 512 | |
| text_mask_padding: true | |
| qk_norm: null | |
| conv_layers: 4 | |
| pe_attn_head: null | |
| attn_backend: torch | |
| attn_mask_enabled: false | |
| checkpoint_activations: false | |
| control: | |
| control_dim: 64 | |
| num_emotions: 6 | |
| null_emotion_id: 0 | |
| neutral_emotion_id: 1 | |
| emotion_embedding_dim: 128 | |
| emotion_hidden_dim: 256 | |
| emotion_depth: 2 | |
| arousal_fourier_dim: 64 | |
| arousal_fourier_scale: 50.0 | |
| arousal_fourier_max_period: 1000.0 | |
| arousal_include_raw: true | |
| arousal_hidden_dim: 256 | |
| arousal_depth: 2 | |
| control_dropout: 0.0 | |
| control_gate_init: 0.1 | |
| control_zero_init: true | |
| injection: | |
| mask: target_only | |
| fusion: add | |
| strategy: axis_split_film_residual | |
| axis_split: | |
| emotion: | |
| film_layers: [11, 17] | |
| hidden_dim: 256 | |
| arousal: | |
| residual_layers: [14, 20] | |
| hidden_dim: 256 | |
| output_gain: 1.0 | |
| time_gate_hidden_dim: 64 | |
| lora: | |
| targets: [to_q, to_k, to_v, to_out.0, ff.0.0, ff.2] | |
| rank: 16 | |
| alpha: 32.0 | |
| dropout: 0.05 | |
| inference: | |
| nfe_steps: 32 | |
| global_guidance: 2.0 | |
| emotion_guidance: 0.0 | |
| arousal_guidance: 0.0 | |
| arousal_scale: 0.75 | |
| sway_sampling_coef: -1.0 | |
| sample_rate: 24000 | |
| hop_length: 256 | |