| { | |
| "version": "1.0", | |
| "truncation": null, | |
| "padding": null, | |
| "added_tokens": [ | |
| { | |
| "id": 0, | |
| "content": "UNK", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 1, | |
| "content": "PAD", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 2, | |
| "content": "WORD_BOUNDARY", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| }, | |
| { | |
| "id": 3, | |
| "content": "UTT_BOUNDARY", | |
| "single_word": false, | |
| "lstrip": false, | |
| "rstrip": false, | |
| "normalized": false, | |
| "special": true | |
| } | |
| ], | |
| "normalizer": { | |
| "type": "Sequence", | |
| "normalizers": [ | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "e-" | |
| }, | |
| "content": "ˈe" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ə-" | |
| }, | |
| "content": "ˈə" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "a-" | |
| }, | |
| "content": "ˈa" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌa" | |
| }, | |
| "content": "ˈa" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌɛ" | |
| }, | |
| "content": "ˈɛ" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌi" | |
| }, | |
| "content": "ˈi" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌɔ" | |
| }, | |
| "content": "ˈɔ" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌɑ̃" | |
| }, | |
| "content": "ˈɑ̃" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˈə-" | |
| }, | |
| "content": "ˈə" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "y-" | |
| }, | |
| "content": "ˈy" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˈe-" | |
| }, | |
| "content": "ˈe" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˈa-" | |
| }, | |
| "content": "ˈa" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌɛ̃" | |
| }, | |
| "content": "ˈɛ̃" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌu" | |
| }, | |
| "content": "ˈu" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌœ" | |
| }, | |
| "content": "ˈœ" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌø" | |
| }, | |
| "content": "ˈø" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌe" | |
| }, | |
| "content": "ˈe" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌə" | |
| }, | |
| "content": "ˈə" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌy" | |
| }, | |
| "content": "ˈy" | |
| }, | |
| { | |
| "type": "Replace", | |
| "pattern": { | |
| "String": "ˌo" | |
| }, | |
| "content": "ˈo" | |
| }, | |
| { | |
| "type": "Strip", | |
| "strip_left": true, | |
| "strip_right": true | |
| } | |
| ] | |
| }, | |
| "pre_tokenizer": { | |
| "type": "WhitespaceSplit" | |
| }, | |
| "post_processor": { | |
| "type": "TemplateProcessing", | |
| "single": [ | |
| { | |
| "SpecialToken": { | |
| "id": "UTT_BOUNDARY", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "A", | |
| "type_id": 0 | |
| } | |
| } | |
| ], | |
| "pair": [ | |
| { | |
| "SpecialToken": { | |
| "id": "UTT_BOUNDARY", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "A", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "SpecialToken": { | |
| "id": "UTT_BOUNDARY", | |
| "type_id": 0 | |
| } | |
| }, | |
| { | |
| "Sequence": { | |
| "id": "B", | |
| "type_id": 1 | |
| } | |
| } | |
| ], | |
| "special_tokens": { | |
| "UTT_BOUNDARY": { | |
| "id": "UTT_BOUNDARY", | |
| "ids": [ | |
| 3 | |
| ], | |
| "tokens": [ | |
| "UTT_BOUNDARY" | |
| ] | |
| } | |
| } | |
| }, | |
| "decoder": null, | |
| "model": { | |
| "type": "WordLevel", | |
| "vocab": { | |
| "UNK": 0, | |
| "PAD": 1, | |
| "WORD_BOUNDARY": 2, | |
| "UTT_BOUNDARY": 3, | |
| "m": 4, | |
| "a": 5, | |
| "ˈɑ̃": 6, | |
| "d": 7, | |
| "ˈɔ": 8, | |
| "n": 9, | |
| "b": 10, | |
| "ˈa": 11, | |
| "ʁ": 12, | |
| "ə": 13, | |
| "ɡ": 14, | |
| "ɔ": 15, | |
| "ʒ": 16, | |
| "ˈi": 17, | |
| "v": 18, | |
| "t": 19, | |
| "k": 20, | |
| "ˈo": 21, | |
| "ˈɛ̃": 22, | |
| "w": 23, | |
| "y": 24, | |
| "j": 25, | |
| "ˈy": 26, | |
| "e": 27, | |
| "ɔ̃": 28, | |
| "ˈe": 29, | |
| "p": 30, | |
| "ɛ": 31, | |
| "f": 32, | |
| "s": 33, | |
| "z": 34, | |
| "l": 35, | |
| "ˈə": 36, | |
| "ˈɛ": 37, | |
| "u": 38, | |
| "o": 39, | |
| "ʃ": 40, | |
| "i": 41, | |
| "ˈu": 42, | |
| "ɛ̃": 43, | |
| "œ": 44, | |
| "ˈø": 45, | |
| "ɑ̃": 46, | |
| "ˈœ": 47, | |
| "ˈɔ̃": 48, | |
| "ø": 49, | |
| "ɲ": 50, | |
| "ˈw": 51, | |
| "t̠ʃ": 52, | |
| "d̠ʒ": 53 | |
| }, | |
| "unk_token": "UNK" | |
| } | |
| } |