ko-ctc-kenlm-spelling-only-wiki / tokenizer_config.json
lIlBrother's picture
Upload 4 files
c73317b
raw
history blame contribute delete
No virus
494 Bytes
{
"bos_token": "<s>",
"do_lower_case": false,
"eos_token": "</s>",
"name_or_path": "/data_raid0/TADEV_BIG_DATA/ASR/STT/model/fine-tuning/spelling/wav2vec2-base-4data",
"pad_token": "<pad>",
"processor_class": "Wav2Vec2ProcessorWithLM",
"replace_word_delimiter_char": " ",
"special_tokens_map_file": "/data/asr_proj/stt/bart_branch/output_dir/grapheme/special_tokens_map.json",
"tokenizer_class": "Wav2Vec2CTCTokenizer",
"unk_token": "<unk>",
"word_delimiter_token": "|"
}