| { |
| "_description": "IPA phone label set used for training. id→symbol mapping from checkpoint phone_to_id.", |
| "_n_classes": 45, |
| "_notes": [ |
| "Class 0 ('sil') covers silence, pauses, and boundary markers (TIMIT h#/pau/epi, Buckeye SIL).", |
| "TIMIT stop closure+release pairs (bcl+b, dcl+d, …) are merged into a single release phone.", |
| "TIMIT 'er/axr' → ɹ̩ (syllabic r); 'el' → l̩; 'em' → m̩; 'en' → n̩.", |
| "TIMIT 'ah/ax/ax-h' → ə (schwa); 'dx' → ɾ (flap); 'q' → ʔ (glottal stop).", |
| "Buckeye nasalised vowel variants (aen, own, …) are merged with their oral counterparts.", |
| "IPA symbol sort order follows Python's default Unicode sort on the training code." |
| ], |
| "id2phone": { |
| "0": "sil", |
| "1": "aɪ", |
| "2": "aʊ", |
| "3": "b", |
| "4": "d", |
| "5": "dʒ", |
| "6": "eɪ", |
| "7": "f", |
| "8": "g", |
| "9": "h", |
| "10": "i", |
| "11": "j", |
| "12": "k", |
| "13": "l", |
| "14": "l̩", |
| "15": "m", |
| "16": "m̩", |
| "17": "n", |
| "18": "n̩", |
| "19": "oʊ", |
| "20": "p", |
| "21": "r", |
| "22": "s", |
| "23": "t", |
| "24": "tʃ", |
| "25": "u", |
| "26": "v", |
| "27": "w", |
| "28": "z", |
| "29": "æ", |
| "30": "ð", |
| "31": "ŋ", |
| "32": "ɑ", |
| "33": "ɔ", |
| "34": "ɔɪ", |
| "35": "ə", |
| "36": "ɛ", |
| "37": "ɪ", |
| "38": "ɹ̩", |
| "39": "ɾ", |
| "40": "ʃ", |
| "41": "ʊ", |
| "42": "ʒ", |
| "43": "ʔ", |
| "44": "θ" |
| }, |
| "phone2id": { |
| "sil": 0, |
| "aɪ": 1, |
| "aʊ": 2, |
| "b": 3, |
| "d": 4, |
| "dʒ": 5, |
| "eɪ": 6, |
| "f": 7, |
| "g": 8, |
| "h": 9, |
| "i": 10, |
| "j": 11, |
| "k": 12, |
| "l": 13, |
| "l̩": 14, |
| "m": 15, |
| "m̩": 16, |
| "n": 17, |
| "n̩": 18, |
| "oʊ": 19, |
| "p": 20, |
| "r": 21, |
| "s": 22, |
| "t": 23, |
| "tʃ": 24, |
| "u": 25, |
| "v": 26, |
| "w": 27, |
| "z": 28, |
| "æ": 29, |
| "ð": 30, |
| "ŋ": 31, |
| "ɑ": 32, |
| "ɔ": 33, |
| "ɔɪ": 34, |
| "ə": 35, |
| "ɛ": 36, |
| "ɪ": 37, |
| "ɹ̩": 38, |
| "ɾ": 39, |
| "ʃ": 40, |
| "ʊ": 41, |
| "ʒ": 42, |
| "ʔ": 43, |
| "θ": 44 |
| } |
| } |
|
|
