mmcauliffe commited on
Commit
48f53c3
·
verified ·
1 Parent(s): a742dac

Upload folder using huggingface_hub

Browse files
README.md CHANGED
@@ -75,7 +75,7 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
75
  #### Details for thai_mfa dictionary and G2P model
76
 
77
  - **Source:** wikipron
78
- - **Orthography:** Latin
79
  - **Phone set:** MFA
80
  - **Words:** 53,840
81
  * **Phones:** 45
@@ -135,7 +135,6 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
135
  - **Male speakers:** 328
136
  - **Unknown speakers:** 5,812
137
 
138
-
139
  #### GlobalPhone Thai
140
 
141
  - **Source:** https://catalogue.elra.info/en-us/repository/browse/ELRA-S0321/
@@ -148,7 +147,6 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
148
  - **Male speakers:** 27
149
  - **Unknown speakers:** 6
150
 
151
-
152
  #### Gowajee Corpus
153
 
154
  - **Source:** https://github.com/ekapolc/gowajee_corpus?tab=readme-ov-file
@@ -161,7 +159,6 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
161
  - **Male speakers:** 163
162
  - **Unknown speakers:** 0
163
 
164
-
165
  #### Lotus Corpus
166
 
167
  - **Source:** https://github.com/korakot/corpus/releases/tag/v1.0
@@ -174,7 +171,6 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
174
  - **Male speakers:** 48
175
  - **Unknown speakers:** 0
176
 
177
-
178
  #### Thai Elderly Speech dataset by Data Wow and VISAI
179
 
180
  - **Source:** https://github.com/VISAI-DATAWOW/Thai-Elderly-Speech-dataset/releases/tag/v1.0.0
@@ -187,7 +183,6 @@ To get started, follow the instructions for [installing MFA](https://montreal-fo
187
  - **Male speakers:** 10
188
  - **Unknown speakers:** 0
189
 
190
-
191
  ### Training Procedure
192
 
193
  <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
 
75
  #### Details for thai_mfa dictionary and G2P model
76
 
77
  - **Source:** wikipron
78
+ - **Orthography:** Thai
79
  - **Phone set:** MFA
80
  - **Words:** 53,840
81
  * **Phones:** 45
 
135
  - **Male speakers:** 328
136
  - **Unknown speakers:** 5,812
137
 
 
138
  #### GlobalPhone Thai
139
 
140
  - **Source:** https://catalogue.elra.info/en-us/repository/browse/ELRA-S0321/
 
147
  - **Male speakers:** 27
148
  - **Unknown speakers:** 6
149
 
 
150
  #### Gowajee Corpus
151
 
152
  - **Source:** https://github.com/ekapolc/gowajee_corpus?tab=readme-ov-file
 
159
  - **Male speakers:** 163
160
  - **Unknown speakers:** 0
161
 
 
162
  #### Lotus Corpus
163
 
164
  - **Source:** https://github.com/korakot/corpus/releases/tag/v1.0
 
171
  - **Male speakers:** 48
172
  - **Unknown speakers:** 0
173
 
 
174
  #### Thai Elderly Speech dataset by Data Wow and VISAI
175
 
176
  - **Source:** https://github.com/VISAI-DATAWOW/Thai-Elderly-Speech-dataset/releases/tag/v1.0.0
 
183
  - **Male speakers:** 10
184
  - **Unknown speakers:** 0
185
 
 
186
  ### Training Procedure
187
 
188
  <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
acoustic/final.alimdl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7ef3f364888d6c2a392d9b734174cef0792e239aec0e425c7cff6bc9af4abc8f
3
- size 50472282
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b20bc3ef53ae796cbf4dd421d3fa8141101c964b31863a01e18ea6b3ae55cb52
3
+ size 50469638
acoustic/final.mdl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:be9cea547459707038b3063dbd9a8d42a7b21f7595a6f788ba1ac4419da62803
3
- size 50472282
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:533994ee7f8ad517521b16502199d00904019791eaae092af69a4ea7757f7241
3
+ size 50469638
acoustic/lda.mat CHANGED
Binary files a/acoustic/lda.mat and b/acoustic/lda.mat differ
 
acoustic/meta.json CHANGED
@@ -1 +1 @@
1
- {"phones": ["aː˥˩", "aː˦˥", "aː˧", "aː˨˩", "aː˩˩˦", "a˥˩", "a˦˥", "a˧", "a˨˩", "a˩˩˦", "b", "d", "eː˥˩", "eː˦˥", "eː˧", "eː˨˩", "eː˩˩˦", "e˥˩", "e˦˥", "e˧", "e˨˩", "e˩˩˦", "f", "h", "ia˥˩", "ia˦˥", "ia˧", "ia˨˩", "ia˩˩˦", "iː˥˩", "iː˦˥", "iː˧", "iː˨˩", "iː˩˩˦", "i˥˩", "i˦˥", "i˧", "i˨˩", "i˩˩˦", "j", "k", "kʰ", "k̚", "l", "m", "n", "oː˥˩", "oː˦˥", "oː˧", "oː˨˩", "oː˩˩˦", "o˥˩", "o˦˥", "o˧", "o˨˩", "o˩˩˦", "p", "pʰ", "p̚", "r", "s", "t", "tɕ", "tɕʰ", "tʰ", "t̚", "ua˥˩", "ua˦˥", "ua˧", "ua˨˩", "ua˩˩˦", "uː˥˩", "uː˦˥", "uː˧", "uː˨˩", "uː˩˩˦", "u˥˩", "u˦˥", "u˧", "u˨˩", "u˩˩˦", "w", "ŋ", "ɔː˥˩", "ɔː˦˥", "ɔː˧", "ɔː˨˩", "ɔː˩˩˦", "ɔ˥˩", "ɔ˦˥", "ɔ˧", "ɔ˨˩", "ɔ˩˩˦", "ɛː˥˩", "ɛː˦˥", "ɛː˧", "ɛː˨˩", "ɛː˩˩˦", "ɛ˥˩", "ɛ˦˥", "ɛ˧", "ɛ˨˩", "ɛ˩˩˦", "ɤː˥˩", "ɤː˦˥", "ɤː˧", "ɤː˨˩", "ɤː˩˩˦", "ɤ˥˩", "ɤ˦˥", "ɤ˧", "ɤ˨˩", "ɤ˩˩˦", "ɯa˥˩", "ɯa˦˥", "ɯa˧", "ɯa˨˩", "ɯa˩˩˦", "ɯː˥˩", "ɯː˦˥", "ɯː˧", "ɯː˨˩", "ɯː˩˩˦", "ɯ˥˩", "ɯ˦˥", "ɯ˧", "ɯ˨˩", "ɯ˩˩˦", "ʔ"], "phone_mapping": {"<eps>": 0, "sil": 1, "spn": 2, "aː˥˩": 3, "aː˦˥": 4, "aː˧": 5, "aː˨˩": 6, "aː˩˩˦": 7, "a˥˩": 8, "a˦˥": 9, "a˧": 10, "a˨˩": 11, "a˩˩˦": 12, "b": 13, "d": 14, "eː˥˩": 15, "eː˦˥": 16, "eː˧": 17, "eː˨˩": 18, "eː˩˩˦": 19, "e˥˩": 20, "e˦˥": 21, "e˧": 22, "e˨˩": 23, "e˩˩˦": 24, "f": 25, "h": 26, "ia˥˩": 27, "ia˦˥": 28, "ia˧": 29, "ia˨˩": 30, "ia˩˩˦": 31, "iː˥˩": 32, "iː˦˥": 33, "iː˧": 34, "iː˨˩": 35, "iː˩˩˦": 36, "i˥˩": 37, "i˦˥": 38, "i˧": 39, "i˨˩": 40, "i˩˩˦": 41, "j": 42, "k": 43, "kʰ": 44, "k̚": 45, "l": 46, "m": 47, "n": 48, "oː˥˩": 49, "oː˦˥": 50, "oː˧": 51, "oː˨˩": 52, "oː˩˩˦": 53, "o˥˩": 54, "o˦˥": 55, "o˧": 56, "o˨˩": 57, "o˩˩˦": 58, "p": 59, "pʰ": 60, "p̚": 61, "r": 62, "s": 63, "t": 64, "tɕ": 65, "tɕʰ": 66, "tʰ": 67, "t̚": 68, "ua˥˩": 69, "ua˦˥": 70, "ua˧": 71, "ua˨˩": 72, "ua˩˩˦": 73, "uː˥˩": 74, "uː˦˥": 75, "uː˧": 76, "uː˨˩": 77, "uː˩˩˦": 78, "u˥˩": 79, "u˦˥": 80, "u˧": 81, "u˨˩": 82, "u˩˩˦": 83, "w": 84, "ŋ": 85, "ɔː˥˩": 86, "ɔː˦˥": 87, "ɔː˧": 88, "ɔː˨˩": 89, "ɔː˩˩˦": 90, "ɔ˥˩": 91, "ɔ˦˥": 92, "ɔ˧": 93, "ɔ˨˩": 94, "ɔ˩˩˦": 95, "ɛː˥˩": 96, "ɛː˦˥": 97, "ɛː˧": 98, "ɛː˨˩": 99, "ɛː˩˩˦": 100, "ɛ˥˩": 101, "ɛ˦˥": 102, "ɛ˧": 103, "ɛ˨˩": 104, "ɛ˩˩˦": 105, "ɤː˥˩": 106, "ɤː˦˥": 107, "ɤː˧": 108, "ɤː˨˩": 109, "ɤː˩˩˦": 110, "ɤ˥˩": 111, "ɤ˦˥": 112, "ɤ˧": 113, "ɤ˨˩": 114, "ɤ˩˩˦": 115, "ɯa˥˩": 116, "ɯa˦˥": 117, "ɯa˧": 118, "ɯa˨˩": 119, "ɯa˩˩˦": 120, "ɯː˥˩": 121, "ɯː˦˥": 122, "ɯː˧": 123, "ɯː˨˩": 124, "ɯː˩˩˦": 125, "ɯ˥˩": 126, "ɯ˦˥": 127, "ɯ˧": 128, "ɯ˨˩": 129, "ɯ˩˩˦": 130, "ʔ": 131}, "phone_groups": {"0": ["aː˥˩", "aː˦˥", "aː˧", "aː˨˩", "aː˩˩˦", "a˥˩", "a˦˥", "a˧", "a˨˩", "a˩˩˦"], "1": ["b"], "2": ["d"], "3": ["eː˥˩", "eː˦˥", "eː˧", "eː˨˩", "eː˩˩˦", "e˥˩", "e˦˥", "e˧", "e˨˩", "e˩˩˦"], "4": ["f"], "5": ["h"], "6": ["ia˥˩", "ia˦˥", "ia˧", "ia˨˩", "ia˩˩˦"], "7": ["iː˥˩", "iː˦˥", "iː˧", "iː˨˩", "iː˩˩˦", "i˥˩", "i˦˥", "i˧", "i˨˩", "i˩˩˦"], "8": ["j"], "9": ["k"], "10": ["kʰ"], "11": ["k̚"], "12": ["l"], "13": ["m"], "14": ["n"], "15": ["oː˥˩", "oː˦˥", "oː˧", "oː˨˩", "oː˩˩˦", "o˥˩", "o˦˥", "o˧", "o˨˩", "o˩˩˦"], "16": ["p"], "17": ["pʰ"], "18": ["p̚"], "19": ["r"], "20": ["s"], "21": ["t"], "22": ["tɕ"], "23": ["tɕʰ"], "24": ["tʰ"], "25": ["t̚"], "26": ["ua˥˩", "ua˦˥", "ua˧", "ua˨˩", "ua˩˩˦"], "27": ["uː˥˩", "uː˦˥", "uː˧", "uː˨˩", "uː˩˩˦", "u˥˩", "u˦˥", "u˧", "u˨˩", "u˩˩˦"], "28": ["w"], "29": ["ŋ"], "30": ["ɔː˥˩", "ɔː˦˥", "ɔː˧", "ɔː˨˩", "ɔː˩˩˦", "ɔ˥˩", "ɔ˦˥", "ɔ˧", "ɔ˨˩", "ɔ˩˩˦"], "31": ["ɛː˥˩", "ɛː˦˥", "ɛː˧", "ɛː˨˩", "ɛː˩˩˦", "ɛ˥˩", "ɛ˦˥", "ɛ˧", "ɛ˨˩", "ɛ˩˩˦"], "32": ["ɤː˥˩", "ɤː˦˥", "ɤː˧", "ɤː˨˩", "ɤː˩˩˦", "ɤ˥˩", "ɤ˦˥", "ɤ˧", "ɤ˨˩", "ɤ˩˩˦"], "33": ["ɯa˥˩", "ɯa˦˥", "ɯa˧", "ɯa˨˩", "ɯa˩˩˦"], "34": ["ɯː˥˩", "ɯː˦˥", "ɯː˧", "ɯː˨˩", "ɯː˩˩˦", "ɯ˥˩", "ɯ˦˥", "ɯ˧", "ɯ˨˩", "ɯ˩˩˦"], "35": ["ʔ"]}, "version": "3.3.0", "architecture": "gmm-hmm", "train_date": "2026-06-10 18:43:44.559682", "training": {"audio_duration": 881098.9370829004, "num_speakers": 6799, "num_utterances": 196286, "num_oovs": 0, "average_log_likelihood": -53.61315001213923}, "dictionaries": {"names": ["thai_mfa"], "default": "thai_mfa", "silence_word": "<eps>", "oov_word": "<unk>", "bracketed_word": "[bracketed]", "laughter_word": "[laughter]", "clitic_marker": null, "position_dependent_phones": false}, "language": "thai", "tokenization": "th", "features": {"type": "mfcc", "use_energy": false, "frame_shift": 10, "frame_length": 25, "snip_edges": false, "low_frequency": 20, "high_frequency": 7800, "sample_frequency": 16000, "dither": 0.0, "energy_floor": 0.0, "num_coefficients": 13, "num_mel_bins": 23, "cepstral_lifter": 22, "preemphasis_coefficient": 0.97, "uses_cmvn": true, "uses_deltas": true, "uses_voiced": false, "uses_splices": false, "uses_speaker_adaptation": true, "use_pitch": true, "use_voicing": true, "min_f0": 50, "max_f0": 800, "delta_pitch": 0.005, "penalty_factor": 0.1, "silence_weight": 0.0, "splice_left_context": 3, "splice_right_context": 3}, "oov_phone": "spn", "optional_silence_phone": "sil", "phone_set_type": "UNKNOWN", "silence_probability": 0.21, "initial_silence_probability": 0.21, "final_silence_correction": 0.99, "final_non_silence_correction": 0.01, "duration_information": {"aː˦˥": [0.16462205455210382, 0.09251521556407023], "aː˧": [0.1279423521209651, 0.07617634794269233], "aː˨˩": [0.11732401018375398, 0.06446537718807539], "aː˩˩˦": [0.1442454634570978, 0.07738156730596822], "a˥˩": [0.04635753547324143, 0.05404549371318417], "a˦˥": [0.06763192869661669, 0.038916796218037966], "a˧": [0.06608249590743781, 0.038990196774968414], "a˨˩": [0.057085443370816426, 0.03157739043677556], "a˩˩˦": [0.06404928104959998, 0.0310690157599518], "b": [0.07322604745872123, 0.027220826859495686], "d": [0.0653299056347771, 0.022944879268525925], "eː˥˩": [0.15099275996624606, 0.08504029830626501], "eː˦˥": [0.1374883692327924, 0.0713840505658281], "eː˧": [0.12696214524444116, 0.07463505984366976], "eː˨˩": [0.11210367531953208, 0.06461165590844786], "eː˩˩˦": [0.13396226507658882, 0.07104169704281726], "e˥˩": [0.0848675230391081, 0.03559666231377905], "e˦˥": [0.06802680005824466, 0.0339691401699723], "e˧": [0.06434866288966905, 0.02879103449296706], "e˨˩": [0.07350385254244396, 0.0338015818244536], "e˩˩˦": [0.07607193603244258, 0.02622792550484669], "i˨˩": [0.06438751353370992, 0.02848934346564786], "i˩˩˦": [0.06935020166520889, 0.0316733626068988], "j": [0.0948610348666577, 0.06174420979906586], "k": [0.08132157390647869, 0.033150487837785805], "kʰ": [0.10309887336645521, 0.03418808183666735], "k̚": [0.08263026316937921, 0.04225585581524279], "l": [0.07220088794085426, 0.03362173121991179], "m": [0.08487747393436657, 0.041589068691366526], "n": [0.08909779091955736, 0.05027804674734586], "oː˥˩": [0.16204810042809942, 0.09088291308053045], "oː˦˥": [0.1433174629060049, 0.0754421659870575], "oː˧": [0.11572049011575918, 0.07318594216594952], "oː˨˩": [0.13449361481308825, 0.06129016272070368], "oː˩˩˦": [0.1509712639719809, 0.12515565316919944], "o˥˩": [0.08404622293779147, 0.03648234872486197], "ua˥˩": [0.09254942140713299, 0.05347535177922171], "ua˦˥": [0.17696831918140343, 0.1067063920322782], "ua˧": [0.1407675310241887, 0.09057831852994842], "ua˨˩": [0.11108795959585172, 0.04574241125478949], "ua˩˩˦": [0.13675726033614524, 0.08285370259465284], "uː˥˩": [0.1125733777386411, 0.0671321610416982], "uː˦˥": [0.1378681186014154, 0.08688639398815601], "uː˧": [0.14173294905361689, 0.08265978252129168], "uː˨˩": [0.11162247024999854, 0.07264192021460779], "uː˩˩˦": [0.15079956577512724, 0.09091275376631333], "u˥˩": [0.07713585520898673, 0.03915384191673356], "ɯ˥˩": [0.08340435256202196, 0.04207149436883967], "aː˥˩": [0.15110903308040516, 0.09614439468278886], "f": [0.11039853766424347, 0.042424535762796395], "h": [0.0930912785400151, 0.03928646912436211], "ia˥˩": [0.13402885013320245, 0.06711286474417213], "ia˦˥": [0.11546575075959506, 0.05258335996311011], "ia˧": [0.1384591491422664, 0.0752006120044877], "ia˨˩": [0.10254485740092556, 0.04980313079896489], "ia˩˩˦": [0.12061705695298458, 0.06611946853194713], "iː˥˩": [0.10307533370051046, 0.0726537727689304], "iː˦˥": [0.22241480016858206, 0.11908577189879252], "iː˧": [0.1517371359565668, 0.1012758439850773], "iː˨˩": [0.13611906942159008, 0.08770379697075972], "iː˩˩˦": [0.1544503284316741, 0.09521164940054841], "i˥˩": [0.08886062101314504, 0.036932452973920664], "i˦˥": [0.06171046372221914, 0.029702729338272876], "i˧": [0.07555496809731597, 0.03870161314650496], "o˦˥": [0.07074536641063381, 0.03535364230473645], "o˧": [0.07507218627136276, 0.03496901698027054], "o˨˩": [0.07271927746934033, 0.033884461707058194], "o˩˩˦": [0.06885181122013631, 0.027936612262581755], "p": [0.06726070024350624, 0.034473059687094396], "pʰ": [0.09003071685479742, 0.03575282990638802], "p̚": [0.08682164004334024, 0.04080383986092763], "r": [0.06749510497081798, 0.03571945889649867], "s": [0.1071171778099476, 0.03753193105129157], "t": [0.07563358682679791, 0.0330296369185516], "tɕ": [0.0827403608381527, 0.027305584308422084], "tɕʰ": [0.11148117291836028, 0.03705770332774204], "tʰ": [0.08627574992063285, 0.032747002590058756], "t̚": [0.08007126846991237, 0.04373385982072948], "ɯ˦˥": [0.06572755315159404, 0.03176344680472012], "ɯ˧": [0.08167735799535335, 0.031099658397526438], "ɯ˨˩": [0.07989874604804431, 0.03496024943294407], "ɯ˩˩˦": [0.07428920600944534, 0.02963550535271697], "ʔ": [0.07650577905457472, 0.035588786395263895], "u˦˥": [0.06186371848307142, 0.032701615737034354], "u˧": [0.061383398036477105, 0.028278672027742792], "u˨˩": [0.0636277642416463, 0.029395519228041185], "u˩˩˦": [0.07354692117462365, 0.030149934670112268], "w": [0.08656866102381308, 0.05444299128023168], "ŋ": [0.09783991777565763, 0.04964327053232243], "ɔː˥˩": [0.12152387462880013, 0.06729115996060069], "ɔː˦˥": [0.14320487400538917, 0.07081486646962369], "ɔː˧": [0.11497282901554118, 0.08424970267817335], "ɔː˨˩": [0.12073429407612656, 0.06273631906005414], "ɔː˩˩˦": [0.09535680616402026, 0.0650537280616926], "ɔ˥˩": [0.08181994121034412, 0.03999864018429338], "ɔ˦˥": [0.08968729085520041, 0.04305753787256104], "ɔ˧": [0.08932258297679245, 0.05751135363988896], "ɔ˨˩": [0.09516308948066828, 0.04390168694259993], "ɔ˩˩˦": [0.08272727843651101, 0.04052828105526045], "ɛː˥˩": [0.14813125158327617, 0.07702765455294036], "ɛː˦˥": [0.1438795460629649, 0.08500309044023603], "ɛː˧": [0.14765057096658818, 0.07921473728239094], "ɛː˨˩": [0.11658944623875503, 0.06489273885400691], "ɛː˩˩˦": [0.14055751262859978, 0.07772922501801939], "ɛ˥˩": [0.10081717929186253, 0.03720874318777794], "ɛ˦˥": [0.07580141331300472, 0.0361198558228694], "ɛ˧": [0.09162079080346892, 0.04666326988737538], "ɛ˨˩": [0.07481421975553476, 0.03282193714816687], "ɛ˩˩˦": [0.08641203188187914, 0.03658401620340452], "ɤː˥˩": [0.15889629171913014, 0.09906048196540222], "ɤː˦˥": [0.15694778702823992, 0.11495581651259582], "ɤː˧": [0.1340000809517729, 0.08247473584838254], "ɤː˨˩": [0.12570360509257703, 0.051698413611527834], "ɤː˩˩˦": [0.18169939083697367, 0.10644193584219894], "ɤ˥˩": [0.11143797708002587, 0.08609741311184789], "ɤ˦˥": [0.1086792472773378, 0.04918111741438924], "ɤ˧": [0.0815121248290049, 0.03447578190715105], "ɤ˨˩": [0.08917695290021936, 0.05345140830549635], "ɤ˩˩˦": [0.11277777949968974, 0.06758635300061137], "ɯa˥˩": [0.10994734047724397, 0.051928566030007645], "ɯa˦˥": [0.18759744496037614, 0.104337197138683], "ɯa˧": [0.14337062637262163, 0.07171807645823731], "ɯa˨˩": [0.13830700946966257, 0.07355486371527127], "ɯa˩˩˦": [0.11674833920570803, 0.0767833011216627], "ɯː˥˩": [0.13956179866473076, 0.07233728254749597], "ɯː˦˥": [0.12017443662747404, 0.07166890197694517], "ɯː˧": [0.13977484625865605, 0.09255119692716364], "ɯː˨˩": [0.12672323549411108, 0.05567442211327999], "ɯː˩˩˦": [0.135472329263798, 0.08724810399151178]}}
 
1
+ {"phones": ["aː˥˩", "aː˦˥", "aː˧", "aː˨˩", "aː˩˩˦", "a˥˩", "a˦˥", "a˧", "a˨˩", "a˩˩˦", "b", "d", "eː˥˩", "eː˦˥", "eː˧", "eː˨˩", "eː˩˩˦", "e˥˩", "e˦˥", "e˧", "e˨˩", "e˩˩˦", "f", "h", "ia˥˩", "ia˦˥", "ia˧", "ia˨˩", "ia˩˩˦", "iː˥˩", "iː˦˥", "iː˧", "iː˨˩", "iː˩˩˦", "i˥˩", "i˦˥", "i˧", "i˨˩", "i˩˩˦", "j", "k", "kʰ", "k̚", "l", "m", "n", "oː˥˩", "oː˦˥", "oː˧", "oː˨˩", "oː˩˩˦", "o˥˩", "o˦˥", "o˧", "o˨˩", "o˩˩˦", "p", "pʰ", "p̚", "r", "s", "t", "tɕ", "tɕʰ", "tʰ", "t̚", "ua˥˩", "ua˦˥", "ua˧", "ua˨˩", "ua˩˩˦", "uː˥˩", "uː˦˥", "uː˧", "uː˨˩", "uː˩˩˦", "u˥˩", "u˦˥", "u˧", "u˨˩", "u˩˩˦", "w", "ŋ", "ɔː˥˩", "ɔː˦˥", "ɔː˧", "ɔː˨˩", "ɔː˩˩˦", "ɔ˥˩", "ɔ˦˥", "ɔ˧", "ɔ˨˩", "ɔ˩˩˦", "ɛː˥˩", "ɛː˦˥", "ɛː˧", "ɛː˨˩", "ɛː˩˩˦", "ɛ˥˩", "ɛ˦˥", "ɛ˧", "ɛ˨˩", "ɛ˩˩˦", "ɤː˥˩", "ɤː˦˥", "ɤː˧", "ɤː˨˩", "ɤː˩˩˦", "ɤ˥˩", "ɤ˦˥", "ɤ˧", "ɤ˨˩", "ɤ˩˩˦", "ɯa˥˩", "ɯa˦˥", "ɯa˧", "ɯa˨˩", "ɯa˩˩˦", "ɯː˥˩", "ɯː˦˥", "ɯː˧", "ɯː˨˩", "ɯː˩˩˦", "ɯ˥˩", "ɯ˦˥", "ɯ˧", "ɯ˨˩", "ɯ˩˩˦", "ʔ"], "phone_mapping": {"<eps>": 0, "sil": 1, "spn": 2, "aː˥˩": 3, "aː˦˥": 4, "aː˧": 5, "aː˨˩": 6, "aː˩˩˦": 7, "a˥˩": 8, "a˦˥": 9, "a˧": 10, "a˨˩": 11, "a˩˩˦": 12, "b": 13, "d": 14, "eː˥˩": 15, "eː˦˥": 16, "eː˧": 17, "eː˨˩": 18, "eː˩˩˦": 19, "e˥˩": 20, "e˦˥": 21, "e˧": 22, "e˨˩": 23, "e˩˩˦": 24, "f": 25, "h": 26, "ia˥˩": 27, "ia˦˥": 28, "ia˧": 29, "ia˨˩": 30, "ia˩˩˦": 31, "iː˥˩": 32, "iː˦˥": 33, "iː˧": 34, "iː˨˩": 35, "iː˩˩˦": 36, "i˥˩": 37, "i˦˥": 38, "i˧": 39, "i˨˩": 40, "i˩˩˦": 41, "j": 42, "k": 43, "kʰ": 44, "k̚": 45, "l": 46, "m": 47, "n": 48, "oː˥˩": 49, "oː˦˥": 50, "oː˧": 51, "oː˨˩": 52, "oː˩˩˦": 53, "o˥˩": 54, "o˦˥": 55, "o˧": 56, "o˨˩": 57, "o˩˩˦": 58, "p": 59, "pʰ": 60, "p̚": 61, "r": 62, "s": 63, "t": 64, "tɕ": 65, "tɕʰ": 66, "tʰ": 67, "t̚": 68, "ua˥˩": 69, "ua˦˥": 70, "ua˧": 71, "ua˨˩": 72, "ua˩˩˦": 73, "uː˥˩": 74, "uː˦˥": 75, "uː˧": 76, "uː˨˩": 77, "uː˩˩˦": 78, "u˥˩": 79, "u˦˥": 80, "u˧": 81, "u˨˩": 82, "u˩˩˦": 83, "w": 84, "ŋ": 85, "ɔː˥˩": 86, "ɔː˦˥": 87, "ɔː˧": 88, "ɔː˨˩": 89, "ɔː˩˩˦": 90, "ɔ˥˩": 91, "ɔ˦˥": 92, "ɔ˧": 93, "ɔ˨˩": 94, "ɔ˩˩˦": 95, "ɛː˥˩": 96, "ɛː˦˥": 97, "ɛː˧": 98, "ɛː˨˩": 99, "ɛː˩˩˦": 100, "ɛ˥˩": 101, "ɛ˦˥": 102, "ɛ˧": 103, "ɛ˨˩": 104, "ɛ˩˩˦": 105, "ɤː˥˩": 106, "ɤː˦˥": 107, "ɤː˧": 108, "ɤː˨˩": 109, "ɤː˩˩˦": 110, "ɤ˥˩": 111, "ɤ˦˥": 112, "ɤ˧": 113, "ɤ˨˩": 114, "ɤ˩˩˦": 115, "ɯa˥˩": 116, "ɯa˦˥": 117, "ɯa˧": 118, "ɯa˨˩": 119, "ɯa˩˩˦": 120, "ɯː˥˩": 121, "ɯː˦˥": 122, "ɯː˧": 123, "ɯː˨˩": 124, "ɯː˩˩˦": 125, "ɯ˥˩": 126, "ɯ˦˥": 127, "ɯ˧": 128, "ɯ˨˩": 129, "ɯ˩˩˦": 130, "ʔ": 131}, "phone_groups": {"0": ["aː˥˩", "aː˦˥", "aː˧", "aː˨˩", "aː˩˩˦", "a˥˩", "a˦˥", "a˧", "a˨˩", "a˩˩˦"], "1": ["b"], "2": ["d"], "3": ["eː˥˩", "eː˦˥", "eː˧", "eː˨˩", "eː˩˩˦", "e˥˩", "e˦˥", "e˧", "e˨˩", "e˩˩˦"], "4": ["f"], "5": ["h"], "6": ["ia˥˩", "ia˦˥", "ia˧", "ia˨˩", "ia˩˩˦"], "7": ["iː˥˩", "iː˦˥", "iː˧", "iː˨˩", "iː˩˩˦", "i˥˩", "i˦˥", "i˧", "i˨˩", "i˩˩˦"], "8": ["j"], "9": ["k"], "10": ["kʰ"], "11": ["k̚"], "12": ["l"], "13": ["m"], "14": ["n"], "15": ["oː˥˩", "oː˦˥", "oː˧", "oː˨˩", "oː˩˩˦", "o˥˩", "o˦˥", "o˧", "o˨˩", "o˩˩˦"], "16": ["p"], "17": ["pʰ"], "18": ["p̚"], "19": ["r"], "20": ["s"], "21": ["t"], "22": ["tɕ"], "23": ["tɕʰ"], "24": ["tʰ"], "25": ["t̚"], "26": ["ua˥˩", "ua˦˥", "ua˧", "ua˨˩", "ua˩˩˦"], "27": ["uː˥˩", "uː˦˥", "uː˧", "uː˨˩", "uː˩˩˦", "u˥˩", "u˦˥", "u˧", "u˨˩", "u˩˩˦"], "28": ["w"], "29": ["ŋ"], "30": ["ɔː˥˩", "ɔː˦˥", "ɔː˧", "ɔː˨˩", "ɔː˩˩˦", "ɔ˥˩", "ɔ˦˥", "ɔ˧", "ɔ˨˩", "ɔ˩˩˦"], "31": ["ɛː˥˩", "ɛː˦˥", "ɛː˧", "ɛː˨˩", "ɛː˩˩˦", "ɛ˥˩", "ɛ˦˥", "ɛ˧", "ɛ˨˩", "ɛ˩˩˦"], "32": ["ɤː˥˩", "ɤː˦˥", "ɤː˧", "ɤː˨˩", "ɤː˩˩˦", "ɤ˥˩", "ɤ˦˥", "ɤ˧", "ɤ˨˩", "ɤ˩˩˦"], "33": ["ɯa˥˩", "ɯa˦˥", "ɯa˧", "ɯa˨˩", "ɯa˩˩˦"], "34": ["ɯː˥˩", "ɯː˦˥", "ɯː˧", "ɯː˨˩", "ɯː˩˩˦", "ɯ˥˩", "ɯ˦˥", "ɯ˧", "ɯ˨˩", "ɯ˩˩˦"], "35": ["ʔ"]}, "version": "3.4.3", "architecture": "gmm-hmm", "train_date": "2026-10-12 13:33:06.942666", "training": {"audio_duration": 880503.5321141428, "num_speakers": 6799, "num_utterances": 196158, "num_oovs": 0, "average_log_likelihood": -54.49728521940001}, "dictionaries": {"names": ["thai_mfa"], "default": "thai_mfa", "silence_word": "<eps>", "oov_word": "<unk>", "bracketed_word": "[bracketed]", "laughter_word": "[laughter]", "clitic_marker": null, "position_dependent_phones": false}, "language": "thai", "tokenization": "th", "features": {"type": "mfcc", "use_energy": false, "frame_shift": 10, "frame_length": 25, "snip_edges": false, "low_frequency": 20, "high_frequency": 7800, "sample_frequency": 16000, "dither": 0.0, "energy_floor": 0.0, "num_coefficients": 13, "num_mel_bins": 23, "cepstral_lifter": 22, "preemphasis_coefficient": 0.97, "uses_cmvn": true, "uses_deltas": true, "uses_voiced": false, "uses_splices": false, "uses_speaker_adaptation": true, "use_pitch": true, "use_voicing": true, "min_f0": 50, "max_f0": 800, "delta_pitch": 0.005, "penalty_factor": 0.1, "silence_weight": 0.0, "splice_left_context": 3, "splice_right_context": 3}, "oov_phone": "spn", "optional_silence_phone": "sil", "phone_set_type": "UNKNOWN", "silence_probability": 0.22, "initial_silence_probability": 0.22, "final_silence_correction": 0.99, "final_non_silence_correction": 0.01, "duration_information": {"aː˨˩": [-2.2927197897242775, 0.538332528290287], "aː˩˩˦": [-2.0829803871961263, 0.5036876181063369], "a˥˩": [-3.4027293747121785, 0.9617614674540338], "a˦˥": [-2.8505610265003485, 0.5226744234040908], "a˧": [-2.881953235208499, 0.5047332523780276], "a˨˩": [-2.9876383110660716, 0.467801209550601], "a˩˩˦": [-2.878430694955748, 0.4423843273486607], "b": [-2.6752923264519595, 0.37043826926889467], "d": [-2.7616829739224174, 0.34325716680146384], "eː˥˩": [-2.059254857191873, 0.5351323392452285], "eː˦˥": [-2.1734554688217513, 0.5541953002046748], "eː˧": [-2.239092182765531, 0.59598555682091], "eː˨˩": [-2.363856611013992, 0.5752487550747648], "eː˩˩˦": [-2.1656888468797395, 0.6138986707206322], "e˥˩": [-2.5525963335876187, 0.4027581554510801], "e˦˥": [-2.7905208864764055, 0.41125682189593515], "e˧": [-2.8216879754655837, 0.4063132167111737], "e˨˩": [-2.733574322966799, 0.3787990733299787], "e˩˩˦": [-2.6714248014921456, 0.3469581693834247], "f": [-2.280351316044919, 0.42297812844870614], "h": [-2.5382931124967922, 0.47744052737026454], "ia˥˩": [-2.128680672816351, 0.46688015860109516], "ia˦˥": [-2.2469929041378385, 0.4071105508986939], "ia˧": [-2.1064843197971865, 0.48350720547569115], "n": [-2.5435601890591593, 0.49926141687591225], "oː˥˩": [-1.9657327165315301, 0.5463906229985546], "oː˦˥": [-2.0741777120888716, 0.5529853249774032], "oː˧": [-2.316958806951139, 0.5842287990036965], "oː˨˩": [-2.1085161274093314, 0.43950930374454805], "oː˩˩˦": [-2.449636328377311, 1.2207381148489953], "o˥˩": [-2.56138620051438, 0.4200877476483584], "o˦˥": [-2.7398748744327452, 0.4480750453298904], "o˧": [-2.682435406274176, 0.43249248494295617], "o˨˩": [-2.7141656818461266, 0.3967032485678466], "o˩˩˦": [-2.754187468945918, 0.3798143800749328], "p": [-2.795394742807972, 0.46560271801114944], "pʰ": [-2.4817171984589588, 0.41520868376035075], "p̚": [-2.561642055407289, 0.46495487715917944], "r": [-2.815240473410524, 0.47382631003523645], "s": [-2.2963885961566977, 0.35622884903081137], "u˧": [-2.875355725490839, 0.40425766112234585], "ɯ˨˩": [-2.6393222522031494, 0.44214252747969823], "ɯ˩˩˦": [-2.7020419059849434, 0.38854093098468195], "ʔ": [-2.685857822696975, 0.43443139475094905], "aː˥˩": [-2.116755097328566, 0.6756205504192805], "aː˦˥": [-1.9820815667078646, 0.5954599343807355], "aː˧": [-2.226505001719766, 0.547949411218116], "ia˨˩": [-2.3926087568867582, 0.468053406767338], "ia˩˩˦": [-2.2345317903986737, 0.46329923645399024], "iː˥˩": [-2.4508878744325258, 0.5690855377693123], "iː˦˥": [-1.686000773198322, 0.5898917949243162], "iː˧": [-2.087293236840569, 0.5929419790138515], "iː˨˩": [-2.20061798616341, 0.6228477972406952], "iː˩˩˦": [-2.0336829826146383, 0.5248881266094655], "i˥˩": [-2.5411230506392983, 0.4140969735354145], "i˦˥": [-2.893746641770424, 0.4699876143574072], "i˧": [-2.7172233138709663, 0.4585639690216852], "i˨˩": [-2.8472143026756394, 0.41901924113857225], "i˩˩˦": [-2.789241490849907, 0.44564081578287396], "j": [-2.533803449169727, 0.548120473587294], "k": [-2.5763561193940223, 0.38973744672600763], "kʰ": [-2.3274345010511737, 0.3643564473493743], "k̚": [-2.612601862096789, 0.45116116418210905], "l": [-2.6932105552208445, 0.413826235306056], "m": [-2.5564655747191245, 0.442390102994893], "t": [-2.646026589449412, 0.41618517196530747], "tɕ": [-2.533559588217683, 0.32958939681360266], "tɕʰ": [-2.2447172279263437, 0.33235805981052696], "tʰ": [-2.517429924662177, 0.4009287452803032], "t̚": [-2.6668357731424206, 0.5013505777844991], "ua˥˩": [-2.5271704643011326, 0.5542399158764248], "ua˦˥": [-1.8752232532597186, 0.5188748299030979], "ua˧": [-2.1276618222800643, 0.5474137500776602], "ua˨˩": [-2.2790225736931924, 0.4085585840579234], "ua˩˩˦": [-2.1554329653854594, 0.541726033000616], "uː˥˩": [-2.320433091388583, 0.5323989709253082], "uː˦˥": [-2.140378716729738, 0.5614186882462174], "uː˧": [-2.100097703579963, 0.540917580221307], "uː˨˩": [-2.3611865570044825, 0.5613324509537125], "uː˩˩˦": [-2.041206212624234, 0.5467808748889293], "u˥˩": [-2.680211946560207, 0.4763335594335503], "u˦˥": [-2.885910614295205, 0.4425509307899769], "u˨˩": [-2.8499480098471732, 0.4176283370023304], "u˩˩˦": [-2.69278395685484, 0.4082999479034582], "w": [-2.587983714644288, 0.5348484662613292], "ŋ": [-2.434397595766789, 0.45994395844671215], "ɔː˥˩": [-2.2446908586172105, 0.498377247512331], "ɔː˦˥": [-2.043386202746574, 0.48612116543310097], "ɔː˧": [-2.4614414471980455, 0.8428004052819257], "ɔː˨˩": [-2.2506583532891558, 0.49535253032578397], "ɔː˩˩˦": [-2.572189865581934, 0.6702521211667306], "ɔ˥˩": [-2.6214914213007083, 0.45125525537269945], "ɔ˦˥": [-2.501848611221241, 0.4841348062207185], "ɔ˧": [-2.604821992026895, 0.569785885430893], "ɔ˨˩": [-2.4308553674637396, 0.4680890921149661], "ɔ˩˩˦": [-2.5956804664101703, 0.4517146179212251], "ɛː˥˩": [-2.0316078590785844, 0.47851593871780945], "ɛː˦˥": [-2.1150055607734686, 0.6850318205378756], "ɛː˧": [-2.0526334652814735, 0.5214797086976726], "ɛː˨˩": [-2.306780373740922, 0.5326935838261345], "ɛː˩˩˦": [-2.1370705490159017, 0.5626554544377157], "ɛ˥˩": [-2.41504285826524, 0.46424906603492644], "ɛ˦˥": [-2.6840954766015286, 0.45045193906603637], "ɛ˧": [-2.5170829468305143, 0.49498587360741964], "ɛ˨˩": [-2.6736695746403703, 0.40853698530449634], "ɛ˩˩˦": [-2.547782677857567, 0.4351733942827578], "ɤː˥˩": [-2.0100443301416737, 0.5625638916160983], "ɤː˦˥": [-2.079701561556576, 0.6241077975342414], "ɤː˧": [-2.1881693671523355, 0.5825437104487379], "ɤː˨˩": [-2.179544849479979, 0.41844922097541726], "ɤː˩˩˦": [-1.8954071356258964, 0.5919034092291621], "ɤ˥˩": [-2.489508971405952, 0.5291483716317931], "ɤ˦˥": [-2.34994394561456, 0.492286059898874], "ɤ˧": [-2.6014358921990657, 0.4151092656250523], "ɤ˨˩": [-2.546120874518283, 0.5534861390729506], "ɤ˩˩˦": [-2.346371135867033, 0.6029039082566126], "ɯa˥˩": [-2.323689661558795, 0.46552924268043316], "ɯa˦˥": [-1.8225458550683882, 0.5209285932193437], "ɯa˧": [-2.0691449039206002, 0.49204102464856275], "ɯa˨˩": [-2.1060498731711323, 0.4833605796802613], "ɯa˩˩˦": [-2.331485828817423, 0.5608915627781069], "ɯː˥˩": [-2.0851397168841657, 0.46765407243982526], "ɯː˦˥": [-2.255301356064994, 0.4973501035141934], "ɯː˧": [-2.15275056053552, 0.5959218642994217], "ɯː˨˩": [-2.1453139249043445, 0.43447066293176406], "ɯː˩˩˦": [-2.1794599965081876, 0.5915322709435318], "ɯ˥˩": [-2.6016750099993335, 0.5010585908218408], "ɯ˦˥": [-2.8302347605098714, 0.45768982127632574], "ɯ˧": [-2.5997791708455757, 0.3826584961405843]}}
acoustic/tree CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c54ba210ca86790dd36b669cd4b4460e25602594cde1920751ffb372ca3c9c1
3
- size 486134
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f4d7c92394a2bcfaeba511cf0ea68d7770acf75a208dddfc5080a7b9fe0c527
3
+ size 477726
dictionary/rules.yaml CHANGED
@@ -1,35 +1,35 @@
1
  dialects:
2
  us:
3
  - following_context: ''
4
- non_silence_before_correction: -0.01
5
  preceding_context: ''
6
  probability: 0.04
7
  replacement: ''
8
  segment: k̚
9
- silence_after_probability: 1.34
10
- silence_before_correction: -0.06
11
  - following_context: ''
12
  non_silence_before_correction: -0.01
13
  preceding_context: ''
14
  probability: 0.02
15
  replacement: ''
16
  segment: p̚
17
- silence_after_probability: 2.58
18
- silence_before_correction: -0.04
19
  - following_context: ''
20
  non_silence_before_correction: -0.01
21
  preceding_context: ''
22
  probability: 0.04
23
  replacement: ''
24
  segment: t̚
25
- silence_after_probability: 1.25
26
- silence_before_correction: -0.04
27
  - following_context: ''
28
  non_silence_before_correction: 0.04
29
  preceding_context: ''
30
  probability: 0.28
31
  replacement: ''
32
  segment: ʔ
33
- silence_after_probability: 0.14
34
- silence_before_correction: -0.2
35
  rules: []
 
1
  dialects:
2
  us:
3
  - following_context: ''
4
+ non_silence_before_correction: -0.03
5
  preceding_context: ''
6
  probability: 0.04
7
  replacement: ''
8
  segment: k̚
9
+ silence_after_probability: 1.43
10
+ silence_before_correction: -0.03
11
  - following_context: ''
12
  non_silence_before_correction: -0.01
13
  preceding_context: ''
14
  probability: 0.02
15
  replacement: ''
16
  segment: p̚
17
+ silence_after_probability: 2.08
18
+ silence_before_correction: -0.03
19
  - following_context: ''
20
  non_silence_before_correction: -0.01
21
  preceding_context: ''
22
  probability: 0.04
23
  replacement: ''
24
  segment: t̚
25
+ silence_after_probability: 1.3
26
+ silence_before_correction: -0.03
27
  - following_context: ''
28
  non_silence_before_correction: 0.04
29
  preceding_context: ''
30
  probability: 0.28
31
  replacement: ''
32
  segment: ʔ
33
+ silence_after_probability: 0.17
34
+ silence_before_correction: -0.18
35
  rules: []
dictionary/thai_mfa.dict CHANGED
The diff for this file is too large to render. See raw diff
 
g2p/thai_mfa/meta.json CHANGED
@@ -1 +1 @@
1
- {"version": "3.2.0", "architecture": "phonetisaurus", "train_date": "2026-05-29 17:16:53.000742", "phones": ["\u025b\u02d0\u02e7", "a\u02d0\u02e9\u02e9\u02e6", "e\u02d0\u02e7", "o\u02e5\u02e9", "p\u02b0", "u\u02d0\u02e8\u02e9", "\u0254\u02e7", "\u014b", "i\u02d0\u02e5\u02e9", "\u0264\u02e9\u02e9\u02e6", "u\u02e5\u02e9", "\u026fa\u02e8\u02e9", "e\u02d0\u02e8\u02e9", "e\u02e6\u02e5", "\u0264\u02e8\u02e9", "k", "o\u02d0\u02e6\u02e5", "\u026fa\u02e7", "l", "\u026f\u02e8\u02e9", "a\u02d0\u02e7", "a\u02d0\u02e6\u02e5", "t\u0255\u02b0", "e\u02e7", "\u025b\u02e8\u02e9", "i\u02e9\u02e9\u02e6", "ia\u02e6\u02e5", "i\u02e6\u02e5", "\u0254\u02d0\u02e7", "\u025b\u02d0\u02e5\u02e9", "a\u02e8\u02e9", "n", "o\u02e9\u02e9\u02e6", "ia\u02e8\u02e9", "k\u031a", "\u0264\u02d0\u02e5\u02e9", "u\u02d0\u02e6\u02e5", "k\u02b0", "e\u02d0\u02e6\u02e5", "o\u02e8\u02e9", "ia\u02e9\u02e9\u02e6", "o\u02e6\u02e5", "r", "ua\u02e6\u02e5", "f", "a\u02e5\u02e9", "t\u0255", "u\u02d0\u02e5\u02e9", "ia\u02e7", "\u025b\u02d0\u02e9\u02e9\u02e6", "b", "p\u031a", "i\u02e5\u02e9", "a\u02e7", "\u0254\u02d0\u02e9\u02e9\u02e6", "o\u02d0\u02e7", "e\u02e9\u02e9\u02e6", "i\u02d0\u02e8\u02e9", "\u025b\u02e5\u02e9", "\u0264\u02d0\u02e7", "e\u02e5\u02e9", "o\u02d0\u02e9\u02e9\u02e6", "o\u02e7", "ua\u02e8\u02e9", "u\u02e9\u02e9\u02e6", "\u026f\u02d0\u02e9\u02e9\u02e6", "p", "w", "ua\u02e9\u02e9\u02e6", "\u0254\u02e8\u02e9", "h", "e\u02e8\u02e9", "\u0264\u02e5\u02e9", "\u025b\u02e9\u02e9\u02e6", "\u026f\u02d0\u02e5\u02e9", "\u026f\u02d0\u02e7", "\u026fa\u02e5\u02e9", "a\u02e9\u02e9\u02e6", "t", "u\u02d0\u02e7", "\u0264\u02e7", "\u0254\u02d0\u02e8\u02e9", "\u026f\u02d0\u02e6\u02e5", "u\u02d0\u02e9\u02e9\u02e6", "\u0264\u02e6\u02e5", "i\u02d0\u02e7", "j", "\u0254\u02e9\u02e9\u02e6", "i\u02d0\u02e6\u02e5", "t\u02b0", "u\u02e7", "\u025b\u02e7", "ua\u02e7", "o\u02d0\u02e5\u02e9", "\u025b\u02d0\u02e6\u02e5", "\u0264\u02d0\u02e9\u02e9\u02e6", "e\u02d0\u02e9\u02e9\u02e6", "i\u02e7", "\u025b\u02e6\u02e5", "a\u02d0\u02e5\u02e9", "\u026f\u02e7", "u\u02e6\u02e5", "s", "\u026f\u02e6\u02e5", "\u0254\u02e6\u02e5", "\u0254\u02d0\u02e6\u02e5", "\u0294", "i\u02e8\u02e9", "\u026f\u02e5\u02e9", "\u026fa\u02e9\u02e9\u02e6", "\u0264\u02d0\u02e6\u02e5", "d", "\u026fa\u02e6\u02e5", "\u0264\u02d0\u02e8\u02e9", "ua\u02e5\u02e9", "ia\u02e5\u02e9", "\u026f\u02d0\u02e8\u02e9", "\u0254\u02e5\u02e9", "m", "a\u02e6\u02e5", "i\u02d0\u02e9\u02e9\u02e6", "u\u02e8\u02e9", "a\u02d0\u02e8\u02e9", "\u026f\u02e9\u02e9\u02e6", "\u0254\u02d0\u02e5\u02e9", "o\u02d0\u02e8\u02e9", "t\u031a", "\u025b\u02d0\u02e8\u02e9", "e\u02d0\u02e5\u02e9"], "graphemes": ["\u0e1c", "\u0e28", "\u0e4d", "\u0e31", "\u0e20", "\u0e09", "\u0e26", "\u0e4a", "\u0e47", "\u0e11", "\u0e36", "\u0e0c", "\u200d", "\u0e06", "\u0e10", "\u0e12", "\u0e1d", "\u0e02", "\u0e22", "\u0e13", "\u0e40", "\u0e0a", "\u0e07", "\u0e03", "\u0e41", "\u0e0e", "\u0e0d", ".", "\u0e1b", "\u0e30", "\u0e49", "\u0e2c", "\u0e08", "\u0e2f", "\u0e18", "\u0e04", "\u0e45", "\u0e4c", "\u0e24", "\u0e0f", "\u0e2e", "\u0e17", "\u0e43", "\u0e1e", "\u0e0b", "\u0e05", "\u0e14", "\u0e29", "\u0e37", "\u0e3a", "\u0e15", "\u0e1a", "\u0e32", "\u0e2d", "\u0e21", "\u0e48", "\u0e35", "\u0e16", "\u0e2b", "\u0e2a", "\u0e19", "\u0e44", "\u0e25", "\u0e38", "\u0e34", "\u0e4b", "\u0e1f", "\u0e01", "\u0e27", "\u0e39", "\u0e42", "\u0e23"], "grapheme_order": 2, "phone_order": 2, "sequence_separator": "|", "unicode_decomposition": false, "evaluation": {"num_words": 3113, "word_error_rate": null, "phone_error_rate": null}, "training": {"num_words": 28006, "num_graphemes": 72, "num_phones": 129}}
 
1
+ {"version": "3.2.0", "architecture": "phonetisaurus", "train_date": "2026-05-29 17:16:53.000742", "phones": ["a\u02e5\u02e9", "\u026f\u02d0\u02e6\u02e5", "e\u02d0\u02e7", "k\u02b0", "\u0264\u02d0\u02e8\u02e9", "e\u02e7", "u\u02d0\u02e5\u02e9", "\u026f\u02e5\u02e9", "k", "u\u02e5\u02e9", "e\u02e9\u02e9\u02e6", "i\u02d0\u02e9\u02e9\u02e6", "\u026f\u02d0\u02e7", "w", "\u026f\u02d0\u02e8\u02e9", "a\u02d0\u02e6\u02e5", "\u0264\u02e8\u02e9", "\u026fa\u02e6\u02e5", "ia\u02e6\u02e5", "s", "p\u02b0", "\u025b\u02d0\u02e6\u02e5", "u\u02d0\u02e9\u02e9\u02e6", "o\u02d0\u02e5\u02e9", "i\u02e5\u02e9", "o\u02e6\u02e5", "ia\u02e7", "\u0264\u02e6\u02e5", "\u0264\u02e7", "ua\u02e6\u02e5", "i\u02e8\u02e9", "\u0264\u02e9\u02e9\u02e6", "a\u02e6\u02e5", "a\u02e9\u02e9\u02e6", "\u0264\u02d0\u02e9\u02e9\u02e6", "ia\u02e9\u02e9\u02e6", "u\u02e7", "t\u0255\u02b0", "u\u02e9\u02e9\u02e6", "a\u02e8\u02e9", "e\u02d0\u02e6\u02e5", "e\u02d0\u02e9\u02e9\u02e6", "e\u02d0\u02e8\u02e9", "a\u02e7", "\u026fa\u02e7", "u\u02d0\u02e6\u02e5", "o\u02e7", "e\u02e5\u02e9", "j", "\u026f\u02d0\u02e9\u02e9\u02e6", "\u0254\u02e5\u02e9", "p\u031a", "\u0254\u02d0\u02e8\u02e9", "\u025b\u02e5\u02e9", "b", "\u0294", "o\u02e8\u02e9", "\u0264\u02d0\u02e5\u02e9", "ua\u02e7", "\u0254\u02d0\u02e5\u02e9", "o\u02d0\u02e9\u02e9\u02e6", "\u0254\u02d0\u02e7", "m", "e\u02d0\u02e5\u02e9", "o\u02d0\u02e7", "\u025b\u02e8\u02e9", "t", "\u0264\u02e5\u02e9", "\u025b\u02d0\u02e7", "\u025b\u02e9\u02e9\u02e6", "\u0264\u02d0\u02e6\u02e5", "\u025b\u02e6\u02e5", "\u025b\u02e7", "p", "\u0254\u02d0\u02e6\u02e5", "r", "a\u02d0\u02e8\u02e9", "\u026f\u02d0\u02e5\u02e9", "o\u02e9\u02e9\u02e6", "k\u031a", "\u025b\u02d0\u02e8\u02e9", "f", "ua\u02e5\u02e9", "\u0254\u02e7", "ua\u02e9\u02e9\u02e6", "ia\u02e8\u02e9", "\u025b\u02d0\u02e5\u02e9", "u\u02d0\u02e8\u02e9", "i\u02d0\u02e6\u02e5", "u\u02e6\u02e5", "i\u02d0\u02e7", "o\u02d0\u02e6\u02e5", "d", "o\u02d0\u02e8\u02e9", "ia\u02e5\u02e9", "\u026f\u02e8\u02e9", "t\u031a", "\u026fa\u02e9\u02e9\u02e6", "t\u02b0", "\u0254\u02e9\u02e9\u02e6", "h", "e\u02e8\u02e9", "\u0264\u02d0\u02e7", "\u026fa\u02e8\u02e9", "i\u02d0\u02e5\u02e9", "i\u02e6\u02e5", "t\u0255", "\u0254\u02e8\u02e9", "\u026f\u02e9\u02e9\u02e6", "\u026fa\u02e5\u02e9", "i\u02e7", "i\u02e9\u02e9\u02e6", "a\u02d0\u02e5\u02e9", "\u026f\u02e6\u02e5", "u\u02e8\u02e9", "o\u02e5\u02e9", "e\u02e6\u02e5", "\u025b\u02d0\u02e9\u02e9\u02e6", "a\u02d0\u02e7", "i\u02d0\u02e8\u02e9", "\u014b", "\u0254\u02d0\u02e9\u02e9\u02e6", "n", "\u0254\u02e6\u02e5", "l", "\u026f\u02e7", "ua\u02e8\u02e9", "u\u02d0\u02e7", "a\u02d0\u02e9\u02e9\u02e6"], "graphemes": ["\u0e42", "\u0e29", "\u0e2d", "\u0e09", "\u0e02", "\u0e20", "\u0e17", "\u0e2e", "\u0e41", "\u0e40", "\u0e19", "\u0e2c", "\u0e04", "\u0e3a", "\u0e39", "\u0e49", "\u0e07", "\u0e03", "\u0e2f", "\u0e22", "\u0e35", "\u0e11", "\u0e37", "\u0e44", "\u0e1b", "\u0e32", "\u0e1d", "\u0e24", "\u0e30", "\u0e14", "\u0e2a", "\u0e4c", "\u0e2b", "\u200d", "\u0e0a", "\u0e25", "\u0e27", "\u0e0c", "\u0e1c", "\u0e0b", "\u0e12", "\u0e16", "\u0e1f", "\u0e13", "\u0e34", "\u0e4d", "\u0e10", "\u0e36", "\u0e05", "\u0e01", ".", "\u0e1e", "\u0e38", "\u0e23", "\u0e06", "\u0e1a", "\u0e28", "\u0e0d", "\u0e0f", "\u0e4b", "\u0e26", "\u0e48", "\u0e47", "\u0e4a", "\u0e45", "\u0e18", "\u0e31", "\u0e43", "\u0e08", "\u0e15", "\u0e0e", "\u0e21"], "grapheme_order": 2, "phone_order": 2, "sequence_separator": "|", "unicode_decomposition": false, "evaluation": {"num_words": 3113, "word_error_rate": null, "phone_error_rate": null}, "training": {"num_words": 28006, "num_graphemes": 72, "num_phones": 129}}