wav2vec_lm_farsi / tokenizer_config.json
tntchack's picture
Upload tokenizer
c5181b7
raw
history blame
492 Bytes
{
"bos_token": null,
"do_lower_case": false,
"eos_token": null,
"model_max_length": 1000000000000000019884624838656,
"name_or_path": "wav2vec_lm_farsi",
"pad_token": "<pad>",
"processor_class": "Wav2Vec2ProcessorWithLM",
"replace_word_delimiter_char": " ",
"special_tokens_map_file": "/workspace/output_models/fa/xlsr-fa/special_tokens_map.json",
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"tokenizer_file": null,
"unk_token": "<unk>",
"word_delimiter_token": "|"
}