Instructions to use facebook/tts_transformer-vi-cv7 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Fairseq
How to use facebook/tts_transformer-vi-cv7 with Fairseq:
from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub models, cfg, task = load_model_ensemble_and_task_from_hf_hub( "facebook/tts_transformer-vi-cv7" ) - Notebooks
- Google Colab
- Kaggle
Download config.yaml from facebook/tts_transformer-vi-cv7: direct link, hf CLI and curl.
- Browser
- Download file 610 Bytes
-
https://huggingface.co/facebook/tts_transformer-vi-cv7/resolve/main/config.yaml
- Command line
-
hf download hf://facebook/tts_transformer-vi-cv7/config.yaml
-
curl -L -o config.yaml https://huggingface.co/facebook/tts_transformer-vi-cv7/resolve/main/config.yaml
610 Bytes
| bpe_tokenizer: | |
| bpe: sentencepiece | |
| sentencepiece_model: spm_char.model | |
| features: | |
| eps: 0.01 | |
| f_max: 8000 | |
| f_min: 20 | |
| hop_len_t: 0.011609977324263039 | |
| hop_length: 256 | |
| n_fft: 1024 | |
| n_mels: 80 | |
| n_stft: 513 | |
| sample_rate: 22050 | |
| type: spectrogram+melscale+log | |
| win_len_t: 0.046439909297052155 | |
| win_length: 1024 | |
| window_fn: hann | |
| global_cmvn: | |
| stats_npz_path: gcmvn_stats.npz | |
| speaker_set_filename: speakers.txt | |
| transforms: | |
| '*': | |
| - global_cmvn | |
| vocab_filename: spm_char.txt | |
| vocoder: | |
| type: hifigan | |
| checkpoint: hifigan.bin | |
| config: hifigan.json | |
| hub: | |
| phonemizer: ipa | |
| lang: vi | |
| speaker: 4191 | |