speech_recog_test / vocab.json
poom-sci's picture
add tokenizer
555d29a
{"m": 0, "i": 1, "j": 2, "b": 3, "d": 4, "v": 5, "'": 6, "e": 7, "q": 8, "n": 9, "s": 10, "c": 11, "l": 12, "u": 13, "t": 14, "k": 15, "f": 16, "o": 17, "w": 18, "p": 19, "h": 20, "y": 21, "g": 23, "r": 24, "z": 25, "a": 26, "x": 27, "|": 22, "[UNK]": 28, "[PAD]": 29}