swedish-common-voice-vox-voxpopuli / tokenizer_config.json
birgermoell's picture
Inital commit
33cc8ab
raw
history blame
181 Bytes
{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}