WAXAL fine-tuning checkpoint for sid: fix pad_token_id + add preprocessor_config.json (from ylacombe/mms-tts-eng-train)
2d6717c verified Download tokenizer_config.json from rnjema-unima/mms-tts-sid-baseline: direct link, hf CLI and curl.
- Browser
- Download file 726 Bytes
-
https://huggingface.co/rnjema-unima/mms-tts-sid-baseline/resolve/main/tokenizer_config.json
- Command line
-
hf download hf://rnjema-unima/mms-tts-sid-baseline/tokenizer_config.json
-
curl -L -o tokenizer_config.json https://huggingface.co/rnjema-unima/mms-tts-sid-baseline/resolve/main/tokenizer_config.json
726 Bytes
| { | |
| "add_blank": true, | |
| "added_tokens_decoder": { | |
| "0": { | |
| "content": "a", | |
| "lstrip": false, | |
| "normalized": false, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": true | |
| }, | |
| "38": { | |
| "content": "<unk>", | |
| "lstrip": false, | |
| "normalized": false, | |
| "rstrip": false, | |
| "single_word": false, | |
| "special": true | |
| } | |
| }, | |
| "backend": "custom", | |
| "clean_up_tokenization_spaces": true, | |
| "is_local": false, | |
| "is_uroman": false, | |
| "language": "sid", | |
| "model_max_length": 1000000000000000019884624838656, | |
| "normalize": true, | |
| "pad_token": "a", | |
| "phonemize": false, | |
| "tokenizer_class": "VitsTokenizer", | |
| "unk_token": "<unk>" | |
| } | |