Automatic Speech Recognition
Transformers
Safetensors
Khmer
English
troryongasr
custom_code
TrorYongASR-small / tokenizer_config.json
Kimang18's picture
Upload processor
62dc600
{
"add_prefix_space": null,
"auto_map": {
"AutoTokenizer": [
"tokenization_troryongasr.TrorYongASRTokenizer",
null
]
},
"backend": "tokenizers",
"bos_token": "<s>",
"clean_up_tokenization_spaces": false,
"eos_token": "</s>",
"extra_special_tokens": [
"<|km|>",
"<|en|>",
"<|transcribe|>",
"<|translate|>",
"<|nospeech|>"
],
"is_local": false,
"model_max_length": 1000000000000000019884624838656,
"padding_side": "right",
"processor_class": "WhisperProcessor",
"sp_model_kwargs": {},
"tokenizer_class": "TrorYongASRTokenizer",
"unk_token": "<unk>",
"use_default_system_prompt": false
}