🚀 Add resample audio when original samplke rate != config[sampling_rate]

Author: dathudeptrai

.gitignore

@@ -37,4 +37,8 @@ dataset/
 mfa/
 kss/
 baker/
-libritts/
+libritts/
+dump_baker/
+dump_ljspeech/
+dump_kss/
+dump_libritts/

tensorflow_tts/bin/preprocess.py

@@ -189,9 +189,11 @@ def gen_audio_features(item, config):
     # check audio properties
     assert len(audio.shape) == 1, f"{utt_id} seems to be multi-channel signal."
     assert np.abs(audio).max() <= 1.0, f"{utt_id} is different from 16 bit PCM."
-    assert (
-        rate == config["sampling_rate"]
-    ), f"{utt_id} sampling rate is not {config['sampling_rate']}."
+    
+    # check sample rate
+    if rate != config["sampling_rate"]:
+        audio = librosa.resample(audio, rate, config["sampling_rate"])
+        logging.info(f"{utt_id} sampling rate is {rate}, not {config['sampling_rate']}, we resample it.")
 
     # trim silence
     if config["trim_silence"]:
@@ -378,7 +380,7 @@ def preprocess():
     processor._save_mapper(
         os.path.join(config["outdir"], f"{config['dataset']}_mapper.json"),
         extra_attrs_to_save={"pinyin_dict": processor.pinyin_dict}
-        if config["dataset"] is "baker"
+        if config["dataset"] == "baker"
         else {},
     )
 

test/files/baker_mapper.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"pad": 0, "sil": 1, "#0": 2, "#1": 3, "#2": 4, "#3": 5, "^": 6, "b": 7, "c": 8, "ch": 9, "d": 10, "f": 11, "g": 12, "h": 13, "j": 14, "k": 15, "l": 16, "m": 17, "n": 18, "p": 19, "q": 20, "r": 21, "s": 22, "sh": 23, "t": 24, "x": 25, "z": 26, "zh": 27, "a1": 28, "a2": 29, "a3": 30, "a4": 31, "a5": 32, "ai1": 33, "ai2": 34, "ai3": 35, "ai4": 36, "ai5": 37, "an1": 38, "an2": 39, "an3": 40, "an4": 41, "an5": 42, "ang1": 43, "ang2": 44, "ang3": 45, "ang4": 46, "ang5": 47, "ao1": 48, "ao2": 49, "ao3": 50, "ao4": 51, "ao5": 52, "e1": 53, "e2": 54, "e3": 55, "e4": 56, "e5": 57, "ei1": 58, "ei2": 59, "ei3": 60, "ei4": 61, "ei5": 62, "en1": 63, "en2": 64, "en3": 65, "en4": 66, "en5": 67, "eng1": 68, "eng2": 69, "eng3": 70, "eng4": 71, "eng5": 72, "er1": 73, "er2": 74, "er3": 75, "er4": 76, "er5": 77, "i1": 78, "i2": 79, "i3": 80, "i4": 81, "i5": 82, "ia1": 83, "ia2": 84, "ia3": 85, "ia4": 86, "ia5": 87, "ian1": 88, "ian2": 89, "ian3": 90, "ian4": 91, "ian5": 92, "iang1": 93, "iang2": 94, "iang3": 95, "iang4": 96, "iang5": 97, "iao1": 98, "iao2": 99, "iao3": 100, "iao4": 101, "iao5": 102, "ie1": 103, "ie2": 104, "ie3": 105, "ie4": 106, "ie5": 107, "ii1": 108, "ii2": 109, "ii3": 110, "ii4": 111, "ii5": 112, "iii1": 113, "iii2": 114, "iii3": 115, "iii4": 116, "iii5": 117, "in1": 118, "in2": 119, "in3": 120, "in4": 121, "in5": 122, "ing1": 123, "ing2": 124, "ing3": 125, "ing4": 126, "ing5": 127, "iong1": 128, "iong2": 129, "iong3": 130, "iong4": 131, "iong5": 132, "iou1": 133, "iou2": 134, "iou3": 135, "iou4": 136, "iou5": 137, "o1": 138, "o2": 139, "o3": 140, "o4": 141, "o5": 142, "ong1": 143, "ong2": 144, "ong3": 145, "ong4": 146, "ong5": 147, "ou1": 148, "ou2": 149, "ou3": 150, "ou4": 151, "ou5": 152, "u1": 153, "u2": 154, "u3": 155, "u4": 156, "u5": 157, "ua1": 158, "ua2": 159, "ua3": 160, "ua4": 161, "ua5": 162, "uai1": 163, "uai2": 164, "uai3": 165, "uai4": 166, "uai5": 167, "uan1": 168, "uan2": 169, "uan3": 170, "uan4": 171, "uan5": 172, "uang1": 173, "uang2": 174, "uang3": 175, "uang4": 176, "uang5": 177, "uei1": 178, "uei2": 179, "uei3": 180, "uei4": 181, "uei5": 182, "uen1": 183, "uen2": 184, "uen3": 185, "uen4": 186, "uen5": 187, "ueng1": 188, "ueng2": 189, "ueng3": 190, "ueng4": 191, "ueng5": 192, "uo1": 193, "uo2": 194, "uo3": 195, "uo4": 196, "uo5": 197, "v1": 198, "v2": 199, "v3": 200, "v4": 201, "v5": 202, "van1": 203, "van2": 204, "van3": 205, "van4": 206, "van5": 207, "ve1": 208, "ve2": 209, "ve3": 210, "ve4": 211, "ve5": 212, "vn1": 213, "vn2": 214, "vn3": 215, "vn4": 216, "vn5": 217, "eos": 218}, "id_to_symbol": {"0": "pad", "1": "sil", "2": "#0", "3": "#1", "4": "#2", "5": "#3", "6": "^", "7": "b", "8": "c", "9": "ch", "10": "d", "11": "f", "12": "g", "13": "h", "14": "j", "15": "k", "16": "l", "17": "m", "18": "n", "19": "p", "20": "q", "21": "r", "22": "s", "23": "sh", "24": "t", "25": "x", "26": "z", "27": "zh", "28": "a1", "29": "a2", "30": "a3", "31": "a4", "32": "a5", "33": "ai1", "34": "ai2", "35": "ai3", "36": "ai4", "37": "ai5", "38": "an1", "39": "an2", "40": "an3", "41": "an4", "42": "an5", "43": "ang1", "44": "ang2", "45": "ang3", "46": "ang4", "47": "ang5", "48": "ao1", "49": "ao2", "50": "ao3", "51": "ao4", "52": "ao5", "53": "e1", "54": "e2", "55": "e3", "56": "e4", "57": "e5", "58": "ei1", "59": "ei2", "60": "ei3", "61": "ei4", "62": "ei5", "63": "en1", "64": "en2", "65": "en3", "66": "en4", "67": "en5", "68": "eng1", "69": "eng2", "70": "eng3", "71": "eng4", "72": "eng5", "73": "er1", "74": "er2", "75": "er3", "76": "er4", "77": "er5", "78": "i1", "79": "i2", "80": "i3", "81": "i4", "82": "i5", "83": "ia1", "84": "ia2", "85": "ia3", "86": "ia4", "87": "ia5", "88": "ian1", "89": "ian2", "90": "ian3", "91": "ian4", "92": "ian5", "93": "iang1", "94": "iang2", "95": "iang3", "96": "iang4", "97": "iang5", "98": "iao1", "99": "iao2", "100": "iao3", "101": "iao4", "102": "iao5", "103": "ie1", "104": "ie2", "105": "ie3", "106": "ie4", "107": "ie5", "108": "ii1", "109": "ii2", "110": "ii3", "111": "ii4", "112": "ii5", "113": "iii1", "114": "iii2", "115": "iii3", "116": "iii4", "117": "iii5", "118": "in1", "119": "in2", "120": "in3", "121": "in4", "122": "in5", "123": "ing1", "124": "ing2", "125": "ing3", "126": "ing4", "127": "ing5", "128": "iong1", "129": "iong2", "130": "iong3", "131": "iong4", "132": "iong5", "133": "iou1", "134": "iou2", "135": "iou3", "136": "iou4", "137": "iou5", "138": "o1", "139": "o2", "140": "o3", "141": "o4", "142": "o5", "143": "ong1", "144": "ong2", "145": "ong3", "146": "ong4", "147": "ong5", "148": "ou1", "149": "ou2", "150": "ou3", "151": "ou4", "152": "ou5", "153": "u1", "154": "u2", "155": "u3", "156": "u4", "157": "u5", "158": "ua1", "159": "ua2", "160": "ua3", "161": "ua4", "162": "ua5", "163": "uai1", "164": "uai2", "165": "uai3", "166": "uai4", "167": "uai5", "168": "uan1", "169": "uan2", "170": "uan3", "171": "uan4", "172": "uan5", "173": "uang1", "174": "uang2", "175": "uang3", "176": "uang4", "177": "uang5", "178": "uei1", "179": "uei2", "180": "uei3", "181": "uei4", "182": "uei5", "183": "uen1", "184": "uen2", "185": "uen3", "186": "uen4", "187": "uen5", "188": "ueng1", "189": "ueng2", "190": "ueng3", "191": "ueng4", "192": "ueng5", "193": "uo1", "194": "uo2", "195": "uo3", "196": "uo4", "197": "uo5", "198": "v1", "199": "v2", "200": "v3", "201": "v4", "202": "v5", "203": "van1", "204": "van2", "205": "van3", "206": "van4", "207": "van5", "208": "ve1", "209": "ve2", "210": "ve3", "211": "ve4", "212": "ve5", "213": "vn1", "214": "vn2", "215": "vn3", "216": "vn4", "217": "vn5", "218": "eos"}, "speakers_map": {"baker": 0}, "processor_name": "BakerProcessor", "pinyin_dict": {"a": ["^", "a"], "ai": ["^", "ai"], "an": ["^", "an"], "ang": ["^", "ang"], "ao": ["^", "ao"], "ba": ["b", "a"], "bai": ["b", "ai"], "ban": ["b", "an"], "bang": ["b", "ang"], "bao": ["b", "ao"], "be": ["b", "e"], "bei": ["b", "ei"], "ben": ["b", "en"], "beng": ["b", "eng"], "bi": ["b", "i"], "bian": ["b", "ian"], "biao": ["b", "iao"], "bie": ["b", "ie"], "bin": ["b", "in"], "bing": ["b", "ing"], "bo": ["b", "o"], "bu": ["b", "u"], "ca": ["c", "a"], "cai": ["c", "ai"], "can": ["c", "an"], "cang": ["c", "ang"], "cao": ["c", "ao"], "ce": ["c", "e"], "cen": ["c", "en"], "ceng": ["c", "eng"], "cha": ["ch", "a"], "chai": ["ch", "ai"], "chan": ["ch", "an"], "chang": ["ch", "ang"], "chao": ["ch", "ao"], "che": ["ch", "e"], "chen": ["ch", "en"], "cheng": ["ch", "eng"], "chi": ["ch", "iii"], "chong": ["ch", "ong"], "chou": ["ch", "ou"], "chu": ["ch", "u"], "chua": ["ch", "ua"], "chuai": ["ch", "uai"], "chuan": ["ch", "uan"], "chuang": ["ch", "uang"], "chui": ["ch", "uei"], "chun": ["ch", "uen"], "chuo": ["ch", "uo"], "ci": ["c", "ii"], "cong": ["c", "ong"], "cou": ["c", "ou"], "cu": ["c", "u"], "cuan": ["c", "uan"], "cui": ["c", "uei"], "cun": ["c", "uen"], "cuo": ["c", "uo"], "da": ["d", "a"], "dai": ["d", "ai"], "dan": ["d", "an"], "dang": ["d", "ang"], "dao": ["d", "ao"], "de": ["d", "e"], "dei": ["d", "ei"], "den": ["d", "en"], "deng": ["d", "eng"], "di": ["d", "i"], "dia": ["d", "ia"], "dian": ["d", "ian"], "diao": ["d", "iao"], "die": ["d", "ie"], "ding": ["d", "ing"], "diu": ["d", "iou"], "dong": ["d", "ong"], "dou": ["d", "ou"], "du": ["d", "u"], "duan": ["d", "uan"], "dui": ["d", "uei"], "dun": ["d", "uen"], "duo": ["d", "uo"], "e": ["^", "e"], "ei": ["^", "ei"], "en": ["^", "en"], "ng": ["^", "en"], "eng": ["^", "eng"], "er": ["^", "er"], "fa": ["f", "a"], "fan": ["f", "an"], "fang": ["f", "ang"], "fei": ["f", "ei"], "fen": ["f", "en"], "feng": ["f", "eng"], "fo": ["f", "o"], "fou": ["f", "ou"], "fu": ["f", "u"], "ga": ["g", "a"], "gai": ["g", "ai"], "gan": ["g", "an"], "gang": ["g", "ang"], "gao": ["g", "ao"], "ge": ["g", "e"], "gei": ["g", "ei"], "gen": ["g", "en"], "geng": ["g", "eng"], "gong": ["g", "ong"], "gou": ["g", "ou"], "gu": ["g", "u"], "gua": ["g", "ua"], "guai": ["g", "uai"], "guan": ["g", "uan"], "guang": ["g", "uang"], "gui": ["g", "uei"], "gun": ["g", "uen"], "guo": ["g", "uo"], "ha": ["h", "a"], "hai": ["h", "ai"], "han": ["h", "an"], "hang": ["h", "ang"], "hao": ["h", "ao"], "he": ["h", "e"], "hei": ["h", "ei"], "hen": ["h", "en"], "heng": ["h", "eng"], "hong": ["h", "ong"], "hou": ["h", "ou"], "hu": ["h", "u"], "hua": ["h", "ua"], "huai": ["h", "uai"], "huan": ["h", "uan"], "huang": ["h", "uang"], "hui": ["h", "uei"], "hun": ["h", "uen"], "huo": ["h", "uo"], "ji": ["j", "i"], "jia": ["j", "ia"], "jian": ["j", "ian"], "jiang": ["j", "iang"], "jiao": ["j", "iao"], "jie": ["j", "ie"], "jin": ["j", "in"], "jing": ["j", "ing"], "jiong": ["j", "iong"], "jiu": ["j", "iou"], "ju": ["j", "v"], "juan": ["j", "van"], "jue": ["j", "ve"], "jun": ["j", "vn"], "ka": ["k", "a"], "kai": ["k", "ai"], "kan": ["k", "an"], "kang": ["k", "ang"], "kao": ["k", "ao"], "ke": ["k", "e"], "kei": ["k", "ei"], "ken": ["k", "en"], "keng": ["k", "eng"], "kong": ["k", "ong"], "kou": ["k", "ou"], "ku": ["k", "u"], "kua": ["k", "ua"], "kuai": ["k", "uai"], "kuan": ["k", "uan"], "kuang": ["k", "uang"], "kui": ["k", "uei"], "kun": ["k", "uen"], "kuo": ["k", "uo"], "la": ["l", "a"], "lai": ["l", "ai"], "lan": ["l", "an"], "lang": ["l", "ang"], "lao": ["l", "ao"], "le": ["l", "e"], "lei": ["l", "ei"], "leng": ["l", "eng"], "li": ["l", "i"], "lia": ["l", "ia"], "lian": ["l", "ian"], "liang": ["l", "iang"], "liao": ["l", "iao"], "lie": ["l", "ie"], "lin": ["l", "in"], "ling": ["l", "ing"], "liu": ["l", "iou"], "lo": ["l", "o"], "long": ["l", "ong"], "lou": ["l", "ou"], "lu": ["l", "u"], "lv": ["l", "v"], "luan": ["l", "uan"], "lve": ["l", "ve"], "lue": ["l", "ve"], "lun": ["l", "uen"], "luo": ["l", "uo"], "ma": ["m", "a"], "mai": ["m", "ai"], "man": ["m", "an"], "mang": ["m", "ang"], "mao": ["m", "ao"], "me": ["m", "e"], "mei": ["m", "ei"], "men": ["m", "en"], "meng": ["m", "eng"], "mi": ["m", "i"], "mian": ["m", "ian"], "miao": ["m", "iao"], "mie": ["m", "ie"], "min": ["m", "in"], "ming": ["m", "ing"], "miu": ["m", "iou"], "mo": ["m", "o"], "mou": ["m", "ou"], "mu": ["m", "u"], "na": ["n", "a"], "nai": ["n", "ai"], "nan": ["n", "an"], "nang": ["n", "ang"], "nao": ["n", "ao"], "ne": ["n", "e"], "nei": ["n", "ei"], "nen": ["n", "en"], "neng": ["n", "eng"], "ni": ["n", "i"], "nia": ["n", "ia"], "nian": ["n", "ian"], "niang": ["n", "iang"], "niao": ["n", "iao"], "nie": ["n", "ie"], "nin": ["n", "in"], "ning": ["n", "ing"], "niu": ["n", "iou"], "nong": ["n", "ong"], "nou": ["n", "ou"], "nu": ["n", "u"], "nv": ["n", "v"], "nuan": ["n", "uan"], "nve": ["n", "ve"], "nue": ["n", "ve"], "nuo": ["n", "uo"], "o": ["^", "o"], "ou": ["^", "ou"], "pa": ["p", "a"], "pai": ["p", "ai"], "pan": ["p", "an"], "pang": ["p", "ang"], "pao": ["p", "ao"], "pe": ["p", "e"], "pei": ["p", "ei"], "pen": ["p", "en"], "peng": ["p", "eng"], "pi": ["p", "i"], "pian": ["p", "ian"], "piao": ["p", "iao"], "pie": ["p", "ie"], "pin": ["p", "in"], "ping": ["p", "ing"], "po": ["p", "o"], "pou": ["p", "ou"], "pu": ["p", "u"], "qi": ["q", "i"], "qia": ["q", "ia"], "qian": ["q", "ian"], "qiang": ["q", "iang"], "qiao": ["q", "iao"], "qie": ["q", "ie"], "qin": ["q", "in"], "qing": ["q", "ing"], "qiong": ["q", "iong"], "qiu": ["q", "iou"], "qu": ["q", "v"], "quan": ["q", "van"], "que": ["q", "ve"], "qun": ["q", "vn"], "ran": ["r", "an"], "rang": ["r", "ang"], "rao": ["r", "ao"], "re": ["r", "e"], "ren": ["r", "en"], "reng": ["r", "eng"], "ri": ["r", "iii"], "rong": ["r", "ong"], "rou": ["r", "ou"], "ru": ["r", "u"], "rua": ["r", "ua"], "ruan": ["r", "uan"], "rui": ["r", "uei"], "run": ["r", "uen"], "ruo": ["r", "uo"], "sa": ["s", "a"], "sai": ["s", "ai"], "san": ["s", "an"], "sang": ["s", "ang"], "sao": ["s", "ao"], "se": ["s", "e"], "sen": ["s", "en"], "seng": ["s", "eng"], "sha": ["sh", "a"], "shai": ["sh", "ai"], "shan": ["sh", "an"], "shang": ["sh", "ang"], "shao": ["sh", "ao"], "she": ["sh", "e"], "shei": ["sh", "ei"], "shen": ["sh", "en"], "sheng": ["sh", "eng"], "shi": ["sh", "iii"], "shou": ["sh", "ou"], "shu": ["sh", "u"], "shua": ["sh", "ua"], "shuai": ["sh", "uai"], "shuan": ["sh", "uan"], "shuang": ["sh", "uang"], "shui": ["sh", "uei"], "shun": ["sh", "uen"], "shuo": ["sh", "uo"], "si": ["s", "ii"], "song": ["s", "ong"], "sou": ["s", "ou"], "su": ["s", "u"], "suan": ["s", "uan"], "sui": ["s", "uei"], "sun": ["s", "uen"], "suo": ["s", "uo"], "ta": ["t", "a"], "tai": ["t", "ai"], "tan": ["t", "an"], "tang": ["t", "ang"], "tao": ["t", "ao"], "te": ["t", "e"], "tei": ["t", "ei"], "teng": ["t", "eng"], "ti": ["t", "i"], "tian": ["t", "ian"], "tiao": ["t", "iao"], "tie": ["t", "ie"], "ting": ["t", "ing"], "tong": ["t", "ong"], "tou": ["t", "ou"], "tu": ["t", "u"], "tuan": ["t", "uan"], "tui": ["t", "uei"], "tun": ["t", "uen"], "tuo": ["t", "uo"], "wa": ["^", "ua"], "wai": ["^", "uai"], "wan": ["^", "uan"], "wang": ["^", "uang"], "wei": ["^", "uei"], "wen": ["^", "uen"], "weng": ["^", "ueng"], "wo": ["^", "uo"], "wu": ["^", "u"], "xi": ["x", "i"], "xia": ["x", "ia"], "xian": ["x", "ian"], "xiang": ["x", "iang"], "xiao": ["x", "iao"], "xie": ["x", "ie"], "xin": ["x", "in"], "xing": ["x", "ing"], "xiong": ["x", "iong"], "xiu": ["x", "iou"], "xu": ["x", "v"], "xuan": ["x", "van"], "xue": ["x", "ve"], "xun": ["x", "vn"], "ya": ["^", "ia"], "yan": ["^", "ian"], "yang": ["^", "iang"], "yao": ["^", "iao"], "ye": ["^", "ie"], "yi": ["^", "i"], "yin": ["^", "in"], "ying": ["^", "ing"], "yo": ["^", "iou"], "yong": ["^", "iong"], "you": ["^", "iou"], "yu": ["^", "v"], "yuan": ["^", "van"], "yue": ["^", "ve"], "yun": ["^", "vn"], "za": ["z", "a"], "zai": ["z", "ai"], "zan": ["z", "an"], "zang": ["z", "ang"], "zao": ["z", "ao"], "ze": ["z", "e"], "zei": ["z", "ei"], "zen": ["z", "en"], "zeng": ["z", "eng"], "zha": ["zh", "a"], "zhai": ["zh", "ai"], "zhan": ["zh", "an"], "zhang": ["zh", "ang"], "zhao": ["zh", "ao"], "zhe": ["zh", "e"], "zhei": ["zh", "ei"], "zhen": ["zh", "en"], "zheng": ["zh", "eng"], "zhi": ["zh", "iii"], "zhong": ["zh", "ong"], "zhou": ["zh", "ou"], "zhu": ["zh", "u"], "zhua": ["zh", "ua"], "zhuai": ["zh", "uai"], "zhuan": ["zh", "uan"], "zhuang": ["zh", "uang"], "zhui": ["zh", "uei"], "zhun": ["zh", "uen"], "zhuo": ["zh", "uo"], "zi": ["z", "ii"], "zong": ["z", "ong"], "zou": ["z", "ou"], "zu": ["z", "u"], "zuan": ["z", "uan"], "zui": ["z", "uei"], "zun": ["z", "uen"], "zuo": ["z", "uo"]}}

test/files/kss_mapper.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"pad": 0, "-": 7, "!": 2, "'": 3, "(": 4, ")": 5, ",": 6, ".": 8, ":": 9, ";": 10, "?": 11, " ": 12, "\u1100": 13, "\u1101": 14, "\u1102": 15, "\u1103": 16, "\u1104": 17, "\u1105": 18, "\u1106": 19, "\u1107": 20, "\u1108": 21, "\u1109": 22, "\u110a": 23, "\u110b": 24, "\u110c": 25, "\u110d": 26, "\u110e": 27, "\u110f": 28, "\u1110": 29, "\u1111": 30, "\u1112": 31, "\u1161": 32, "\u1162": 33, "\u1163": 34, "\u1164": 35, "\u1165": 36, "\u1166": 37, "\u1167": 38, "\u1168": 39, "\u1169": 40, "\u116a": 41, "\u116b": 42, "\u116c": 43, "\u116d": 44, "\u116e": 45, "\u116f": 46, "\u1170": 47, "\u1171": 48, "\u1172": 49, "\u1173": 50, "\u1174": 51, "\u1175": 52, "\u11a8": 53, "\u11a9": 54, "\u11aa": 55, "\u11ab": 56, "\u11ac": 57, "\u11ad": 58, "\u11ae": 59, "\u11af": 60, "\u11b0": 61, "\u11b1": 62, "\u11b2": 63, "\u11b3": 64, "\u11b4": 65, "\u11b5": 66, "\u11b6": 67, "\u11b7": 68, "\u11b8": 69, "\u11b9": 70, "\u11ba": 71, "\u11bb": 72, "\u11bc": 73, "\u11bd": 74, "\u11be": 75, "\u11bf": 76, "\u11c0": 77, "\u11c1": 78, "\u11c2": 79, "eos": 80}, "id_to_symbol": {"0": "pad", "1": "-", "2": "!", "3": "'", "4": "(", "5": ")", "6": ",", "7": "-", "8": ".", "9": ":", "10": ";", "11": "?", "12": " ", "13": "\u1100", "14": "\u1101", "15": "\u1102", "16": "\u1103", "17": "\u1104", "18": "\u1105", "19": "\u1106", "20": "\u1107", "21": "\u1108", "22": "\u1109", "23": "\u110a", "24": "\u110b", "25": "\u110c", "26": "\u110d", "27": "\u110e", "28": "\u110f", "29": "\u1110", "30": "\u1111", "31": "\u1112", "32": "\u1161", "33": "\u1162", "34": "\u1163", "35": "\u1164", "36": "\u1165", "37": "\u1166", "38": "\u1167", "39": "\u1168", "40": "\u1169", "41": "\u116a", "42": "\u116b", "43": "\u116c", "44": "\u116d", "45": "\u116e", "46": "\u116f", "47": "\u1170", "48": "\u1171", "49": "\u1172", "50": "\u1173", "51": "\u1174", "52": "\u1175", "53": "\u11a8", "54": "\u11a9", "55": "\u11aa", "56": "\u11ab", "57": "\u11ac", "58": "\u11ad", "59": "\u11ae", "60": "\u11af", "61": "\u11b0", "62": "\u11b1", "63": "\u11b2", "64": "\u11b3", "65": "\u11b4", "66": "\u11b5", "67": "\u11b6", "68": "\u11b7", "69": "\u11b8", "70": "\u11b9", "71": "\u11ba", "72": "\u11bb", "73": "\u11bc", "74": "\u11bd", "75": "\u11be", "76": "\u11bf", "77": "\u11c0", "78": "\u11c1", "79": "\u11c2", "80": "eos"}, "speakers_map": {"kss": 0}, "processor_name": "KSSProcessor"}

test/files/libritts_mapper.json

@@ -0,0 +1 @@
+{"symbol_to_id": {"@<pad>": 0, "@<unk>": 1, "@<s>": 2, "@</s>": 3, "@AA0": 4, "@AA1": 5, "@AA2": 6, "@AE0": 7, "@AE1": 8, "@AE2": 9, "@AH0": 10, "@AH1": 11, "@AH2": 12, "@AO0": 13, "@AO1": 14, "@AO2": 15, "@AW0": 16, "@AW1": 17, "@AW2": 18, "@AY0": 19, "@AY1": 20, "@AY2": 21, "@B": 22, "@CH": 23, "@D": 24, "@DH": 25, "@EH0": 26, "@EH1": 27, "@EH2": 28, "@ER0": 29, "@ER1": 30, "@ER2": 31, "@EY0": 32, "@EY1": 33, "@EY2": 34, "@F": 35, "@G": 36, "@HH": 37, "@IH0": 38, "@IH1": 39, "@IH2": 40, "@IY0": 41, "@IY1": 42, "@IY2": 43, "@JH": 44, "@K": 45, "@L": 46, "@M": 47, "@N": 48, "@NG": 49, "@OW0": 50, "@OW1": 51, "@OW2": 52, "@OY0": 53, "@OY1": 54, "@OY2": 55, "@P": 56, "@R": 57, "@S": 58, "@SH": 59, "@T": 60, "@TH": 61, "@UH0": 62, "@UH1": 63, "@UH2": 64, "@UW": 65, "@UW0": 66, "@UW1": 67, "@UW2": 68, "@V": 69, "@W": 70, "@Y": 71, "@Z": 72, "@ZH": 73, "@SIL": 74, "@END": 75, "!": 76, "'": 77, "(": 78, ")": 79, ",": 80, ".": 81, ":": 82, ";": 83, "?": 84, " ": 85}, "id_to_symbol": {"0": "@<pad>", "1": "@<unk>", "2": "@<s>", "3": "@</s>", "4": "@AA0", "5": "@AA1", "6": "@AA2", "7": "@AE0", "8": "@AE1", "9": "@AE2", "10": "@AH0", "11": "@AH1", "12": "@AH2", "13": "@AO0", "14": "@AO1", "15": "@AO2", "16": "@AW0", "17": "@AW1", "18": "@AW2", "19": "@AY0", "20": "@AY1", "21": "@AY2", "22": "@B", "23": "@CH", "24": "@D", "25": "@DH", "26": "@EH0", "27": "@EH1", "28": "@EH2", "29": "@ER0", "30": "@ER1", "31": "@ER2", "32": "@EY0", "33": "@EY1", "34": "@EY2", "35": "@F", "36": "@G", "37": "@HH", "38": "@IH0", "39": "@IH1", "40": "@IH2", "41": "@IY0", "42": "@IY1", "43": "@IY2", "44": "@JH", "45": "@K", "46": "@L", "47": "@M", "48": "@N", "49": "@NG", "50": "@OW0", "51": "@OW1", "52": "@OW2", "53": "@OY0", "54": "@OY1", "55": "@OY2", "56": "@P", "57": "@R", "58": "@S", "59": "@SH", "60": "@T", "61": "@TH", "62": "@UH0", "63": "@UH1", "64": "@UH2", "65": "@UW", "66": "@UW0", "67": "@UW1", "68": "@UW2", "69": "@V", "70": "@W", "71": "@Y", "72": "@Z", "73": "@ZH", "74": "@SIL", "75": "@END", "76": "!", "77": "'", "78": "(", "79": ")", "80": ",", "81": ".", "82": ":", "83": ";", "84": "?", "85": " "}, "speakers_map": {"200": 0, "1841": 1, "3664": 2, "6454": 3, "8108": 4, "2416": 5, "4680": 6, "6147": 7, "412": 8, "2952": 9, "8838": 10, "2836": 11, "1263": 12, "5322": 13, "3830": 14, "7447": 15, "1116": 16, "8312": 17, "8123": 18, "250": 19}, "processor_name": "LibriTTSProcessor"}