new_ipa_wav2vec2_timit / tokenizer_config.json
ppakji's picture
add tokenizer
d342655
raw
history blame contribute delete
263 Bytes
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "do_lower_case": false, "word_delimiter_token": "|", "do_phonemize": true, "phonemizer_lang": "en-us", "phonemizer_backend": "espeak", "tokenizer_class": "Wav2Vec2CTCTokenizer"}