Upload Symphonym v7 model, vocabularies, and evaluation results

README.md

@@ -0,0 +1,251 @@
+---
+language:
+- multilingual
+- ar
+- zh
+- ru
+- ja
+- ko
+- he
+- fa
+- hi
+- el
+- ka
+- am
+- hy
+- ur
+- bn
+- ta
+- te
+- th
+tags:
+- toponym-matching
+- cross-script
+- phonetic-embeddings
+- geospatial
+- named-entity
+- information-retrieval
+- teacher-student
+- knowledge-distillation
+license: cc-by-4.0
+datasets:
+- geonames
+- wikidata
+- getty-tgn
+metrics:
+- recall@k
+- mrr
+pipeline_tag: feature-extraction
+---
+
+# Symphonym v7 — Universal Phonetic Embeddings for Cross-Script Toponym Matching
+
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.18682017.svg)](https://doi.org/10.5281/zenodo.18682017)
+
+Symphonym maps toponyms (place names) from **20 writing systems** into a unified
+**128-dimensional phonetic embedding space**, enabling direct cross-script similarity
+comparison without runtime phonetic conversion or language identification.
+
+> *"London" / "Лондон" / "伦敦" / "لندن" → [0.12, -0.34, …] (all nearby)*
+
+## Intended Use
+
+- **Cross-script toponym matching** in geographic databases and gazetteers
+- **Phonetic search** — retrieve results for a place name entered in any script
+- **Historical record linkage** — match pre-standardisation spelling variants
+- **Multilingual named entity linking** in NLP pipelines
+- **Digital humanities** — reconciling place references across archival sources
+
+The model operates on **phonetic similarity**, not semantic or orthographic similarity.
+It is designed as a **candidate retrieval** component within a larger reconciliation
+pipeline, where candidates are subsequently filtered by geographic proximity and
+other constraints.
+
+## Quick Start
+
+```python
+from inference import SymphonymModel
+
+model = SymphonymModel()   # loads weights from this directory
+
+# Single similarity score
+sim = model.similarity("London", "en", "Лондон", "ru")
+print(f"London / Лондон: {sim:.3f}")   # → 0.991
+
+# Batch embeddings  (N × 128 numpy array)
+embeddings = model.batch_embed([
+    ("London",   "en"),
+    ("Лондон",   "ru"),
+    ("伦敦",     "zh"),
+    ("لندن",     "ar"),
+    ("ლონდონი",  "ka"),
+])
+```
+
+### With HuggingFace `huggingface_hub`
+
+```python
+from huggingface_hub import snapshot_download
+
+model_dir = snapshot_download("docuracy/symphonym-v7")
+
+from inference import SymphonymModel
+model = SymphonymModel(model_dir=model_dir)
+```
+
+## Representative Cross-Script Similarities
+
+| Pair | Scripts | Similarity |
+|------|---------|-----------|
+| London / Лондон | Latin–Cyrillic | 0.991 |
+| Athens / Αθήνα | Latin–Greek | 0.980 |
+| Beijing / 北京 | Latin–CJK | 0.955 |
+| Baghdad / بغداد | Latin–Arabic | 0.969 |
+| Jerusalem / ירושלים | Latin–Hebrew | 0.892 |
+| Tokyo / とうきょう | Latin–Hiragana | ~0.94 |
+| London / Londres | Latin–Latin | 0.474 *(correct: phonetically distinct)* |
+
+## Model Architecture
+
+Symphonym uses a **Teacher–Student knowledge distillation** framework.
+
+### Teacher (PhoneticEncoder) — training only
+- Input: IPA transcriptions via Epitran (+ 102 extensions), Phonikud, CharsiuG2P
+- Representation: PanPhon192 — 24-dim articulatory feature vectors,
+  8-bin positional pooling → 192-dim fixed-length input
+- Architecture: BiLSTM → Self-Attention → Attention Pooling → 128-dim projection
+
+### Student (UniversalEncoder) — deployed model
+- Input: raw Unicode characters + script ID + language ID + length bucket
+- Vocabulary: **113,280 tokens** across 20 scripts, 1,944 language codes
+- Architecture: Character/Script/Language/Length embeddings →
+  Input projection → BiLSTM → Self-Attention (residual) →
+  Attention Pooling → 128-dim projection → L2 normalisation
+- Parameters: ~8.3M
+
+The **length bucket embedding** (16 buckets, 8-dim) conditions every character
+representation on sequence length, mitigating spurious matches between short
+toponyms and long compound strings.
+
+### Three-Phase Training Curriculum
+
+| Phase | Objective | Epochs | Notes |
+|-------|-----------|--------|-------|
+| 1 | Teacher: triplet margin loss on PanPhon192 features | 50 | val_loss 0.0056 |
+| 2 | Student–Teacher distillation: α·MSE + (1−α)·cosine | 50 | α=0.5, Student-Teacher cosine 0.942 |
+| 3 | Hard negative fine-tuning (triplet, margin=0.3) | 30 | val_loss 0.02122 |
+
+## Evaluation
+
+### MEHDIE Hebrew–Arabic Historical Benchmark (Sagi et al., 2025)
+
+Independent evaluation on medieval Hebrew and Arabic geographical sources — **not in training data**.
+
+| Method | R@1 | R@5 | R@10 | MRR |
+|--------|-----|-----|------|-----|
+| PanPhon192 (ablation) | 41.1% | 48.2% | 52.3% | 45.0% |
+| Levenshtein + AnyAscii | 81.5% | 97.5% | 99.4% | 88.5% |
+| Jaro-Winkler + AnyAscii | 78.5% | 96.2% | 97.8% | 86.3% |
+| **Symphonym v7** | **85.2%** | **97.0%** | **97.6%** | **90.8%** |
+
+The PanPhon192 ablation (raw articulatory features, no neural training) achieves only
+45.0% MRR — less than half Symphonym's score and below the string baselines —
+confirming that performance derives from the training curriculum, not the phonetic
+features alone.
+
+### Cross-Script Pair Validation (11,723 pairs, 170+ script combinations)
+
+Systematically sampled from training data (up to 10 pairs per script-pair bin); these
+test embedding retrieval quality over the full 67M-toponym index, not generalisation
+to unseen sources.
+
+| Metric | v6 | v7 |
+|--------|----|----|
+| Pass rate (≥0.75 cosine) | — | **90.7%** |
+| Embedding coverage | ~98% | **100%** |
+| Hiragana↔Katakana mean similarity | 0.000 | **0.981** |
+
+**Best-performing script pairs:** Hiragana–Katakana (0.981), Devanagari–Kannada (0.976),
+Devanagari–Telugu (0.976), Cyrillic–Latin (0.923, n=1,334), Arabic–Latin (0.898, n=800).
+
+## v7 Changes
+
+v6 exhibited 0% IPA coverage for Hiragana (151,980 toponyms) and Katakana (340,555 toponyms)
+despite both being natively supported by Epitran (`jpn-Hira`, `jpn-Kana`). The pipeline
+was dispatching by language first (`lang=ja`), routing all Japanese toponyms to CharsiuG2P
+which only processes CJK/Kanji. v7 fixes this by dispatching on detected script *before*
+language code, restoring IPA coverage for 492,535 toponyms. The model was retrained from scratch.
+
+## Training Data
+
+Trained on 66.9 million unique toponyms from:
+
+| Source | License |
+|--------|---------|
+| GeoNames | CC BY 4.0 |
+| Wikidata | CC0 |
+| Getty TGN | ODC-By 1.0 |
+
+54.0% of training-namespace toponyms received IPA transcription; the remainder
+contribute to the Student's character-level learning via distillation.
+
+## Repository Contents
+
+```
+model.safetensors          Student (UniversalEncoder) weights
+config.json                Architecture hyperparameters
+inference.py               Self-contained inference module
+requirements.txt           Dependencies
+vocab/
+  char_vocab.json          113,280-character vocabulary
+  lang_vocab.json          1,944 ISO language codes
+  script_vocab.json        20 script categories
+evaluation/
+  mehdie_results_v7_ranking.json
+  symphonym_v7_pairs_test_report.json
+training_stats/
+  coverage_stats.json      IPA coverage by script and language
+  phase{1,2,3}_metrics.json
+epitran_extensions/        102 custom CSV G2P files
+```
+
+## Limitations
+
+- **Phonetic similarity only**: The model does not use geographic coordinates,
+  semantic information, or entity types. Phonetically similar but geographically
+  unrelated names (Austria/Australia: 0.883) will score highly.
+- **Training bias**: Sources over-represent populated places with official names
+  in high-resource languages. Performance on under-represented scripts and
+  mundane places may be weaker.
+- **Tonal languages**: PanPhon encodes segmental articulatory features but not
+  tone. Tonal minimal pairs in place names are rare in practice.
+- **CJK–Hiragana pairs**: Mean similarity 0.437, reflecting that CharsiuG2P
+  produces Mandarin phonetics for Kanji while Epitran produces Japanese readings
+  for Hiragana — a genuine phonological mismatch, not a model deficiency.
+
+## Citation
+
+If you use Symphonym in your research, please cite the preprint and the Zenodo dataset:
+
+```bibtex
+@misc{symphonym2025,
+    author       = {Gadd, Stephen},
+    title        = {Symphonym: Universal Phonetic Embeddings for Cross-Script Name Matching},
+    year         = {2026},
+    eprint       = {2601.06932},
+    archivePrefix = {arXiv},
+    primaryClass = {cs.CL},
+    url          = {https://arxiv.org/abs/2601.06932},
+    doi          = {10.48550/arXiv.2601.06932}
+}
+
+@dataset{symphonym_v7_zenodo,
+    title   = {Symphonym v7 — Universal Phonetic Embeddings for Cross-Script Toponym Matching},
+    year    = {2026},
+    doi     = {10.5281/zenodo.18682017},
+    url     = {https://doi.org/10.5281/zenodo.18682017}
+}
+```
+
+
+

config.json

@@ -0,0 +1,76 @@
+{
+  "model_type": "symphonym",
+  "architectures": ["UniversalEncoder"],
+  "version": "v7",
+
+  "vocab_size": 113280,
+  "num_scripts": 25,
+  "num_langs": 1944,
+  "num_length_buckets": 16,
+
+  "char_embed_dim": 64,
+  "script_embed_dim": 16,
+  "lang_embed_dim": 16,
+  "length_embed_dim": 8,
+  "hidden_dim": 128,
+  "embed_dim": 128,
+  "num_layers": 2,
+  "num_attention_heads": 2,
+  "dropout": 0.2,
+  "lang_dropout": 0.5,
+
+  "max_toponym_len": 50,
+
+  "teacher": {
+    "architectures": ["PhoneticEncoder"],
+    "phonetic_feat_dim": 192,
+    "hidden_dim": 128,
+    "embed_dim": 128,
+    "num_layers": 2,
+    "num_attention_heads": 2,
+    "dropout": 0.2,
+    "note": "PanPhon192: 8-bin positional pooling of 24-dim articulatory features"
+  },
+
+  "training": {
+    "phase1_epochs": 50,
+    "phase2_epochs": 50,
+    "phase3_epochs": 30,
+    "phase1_val_loss": 0.0056,
+    "phase2_student_teacher_cosine": 0.942,
+    "phase3_val_loss": 0.021217376736320875,
+    "triplet_margin": 0.3,
+    "phase1_lr": 1e-4,
+    "phase2_lr": 5e-4,
+    "phase3_lr": 5e-5,
+    "phase1_batch_size": 2048,
+    "phase2_batch_size": 1024,
+    "phase3_batch_size": 1024,
+    "weight_decay": 1e-5,
+    "warmup_epochs": 2,
+    "noise_prob": 0.3,
+    "mse_weight": 1.0,
+    "cosine_weight": 1.0,
+    "phase2_alpha": 0.5,
+    "phase2_loss": "alpha * MSE(student, teacher) + (1 - alpha) * (1 - cosine(student, teacher))",
+    "phase3_loss": "triplet_margin(anchor, positive, hard_negative, margin=0.3)"
+  },
+
+  "g2p": {
+    "backends": ["epitran", "phonikud", "charsiu_g2p"],
+    "epitran_extensions": 102,
+    "scripts_covered": 20,
+    "ipa_coverage": 0.540,
+    "panphon_bins": 8,
+    "panphon_feat_dim": 24,
+    "panphon192_dim": 192
+  },
+
+  "index": {
+    "total_toponyms": 66924548,
+    "embedding_coverage": 1.0,
+    "quantization": "int8",
+    "ann_backend": "elasticsearch_hnsw"
+  }
+}
+

epitran_extensions/ace-Latn.csv

@@ -0,0 +1,35 @@
+Orth,Phon
+bh,b
+dh,d
+eu,ɤ
+ng,ŋ
+ph,p
+th,t
+a,a
+b,b
+c,tʃ
+d,d
+e,ə
+é,e
+ë,ə
+f,f
+g,ɡ
+h,h
+i,i
+j,ɟ
+k,k
+l,l
+m,m
+n,n
+o,ɔ
+ô,o
+p,p
+q,ʔ
+r,r
+s,s
+t,t
+u,u
+w,w
+x,ks
+y,j
+z,z

epitran_extensions/afr-Latn.csv

@@ -0,0 +1,99 @@
+Orth,Phon
+a,ɑ
+â,ɑː
+aa,ɑː
+ae,ɑɛ
+ai,ɑi
+au,ɑu
+b,b
+c,k
+ch,ʃ
+d,d
+dj,dʒ
+e,ɛ
+ê,eː
+ee,eː
+ei,ɛi
+eu,øː
+f,f
+g,x
+gh,ɣ
+h,ɦ
+i,i
+î,əː
+ie,i
+j,j
+k,k
+l,l
+m,m
+n,n
+ng,ŋ
+o,ɔ
+ô,oː
+oe,u
+oo,oː
+ooi,oːi
+ou,ɑu
+p,p
+r,r
+s,s
+sj,ʃ
+t,t
+tj,tʃ
+u,œ
+û,œː
+ui,œy
+uu,yː
+v,f
+w,v
+y,ɛi
+z,s
+A,ɑ
+Â,ɑː
+Aa,ɑː
+Ae,ɑɛ
+Ai,ɑi
+Au,ɑu
+B,b
+C,k
+Ch,ʃ
+D,d
+Dj,dʒ
+E,ɛ
+Ê,eː
+Ee,eː
+Ei,ɛi
+Eu,øː
+F,f
+G,x
+Gh,ɣ
+H,ɦ
+I,i
+Î,əː
+Ie,i
+J,j
+K,k
+L,l
+M,m
+N,n
+Ng,ŋ
+O,ɔ
+Ô,oː
+Oe,u
+Oo,oː
+Ooi,oːi
+Ou,ɑu
+P,p
+R,r
+S,s
+Sj,ʃ
+T,t
+Tj,tʃ
+U,œ
+Û,œː
+Ui,œy
+Uu,yː
+V,f
+W,v
+Y,ɛi
+Z,s

epitran_extensions/arg-Latn.csv

@@ -0,0 +1,72 @@
+Orth,Phon
+ch,tʃ
+ll,ʎ
+ny,ɲ
+gu,ɡ
+qu,k
+ix,jʃ
+tx,tʃ
+ñ,ɲ
+Ñ,ɲ
+á,a
+é,e
+í,i
+ó,o
+ú,u
+à,a
+è,ɛ
+ì,i
+ò,ɔ
+ù,u
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,
+i,i
+j,tʃ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,ɾ
+s,s
+t,t
+u,u
+v,b
+w,w
+x,ʃ
+y,j
+z,θ
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,
+I,i
+J,tʃ
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,ɾ
+S,s
+T,t
+U,u
+V,b
+W,w
+X,ʃ
+Y,j
+Z,θ

epitran_extensions/arz-Arab.csv

@@ -0,0 +1,73 @@
+Orth,Phon
+الش,aʃʃ
+الس,ass
+الص,asˤsˤ
+الض,adˤdˤ
+الط,atˤtˤ
+الظ,azˤzˤ
+الل,all
+الن,ann
+الر,arr
+الز,azz
+الد,add
+الذ,azz
+الت,att
+الث,ass
+ال,al
+يّ,ijj
+وّ,uww
+ية,ijja
+ات,aːt
+ين,iːn
+او,oː
+اي,eː
+وا,uː
+يا,jaː
+آ,ʔaː
+أ,ʔa
+إ,ʔi
+ى,aː
+ة,a
+ﺔ,a
+ا,aː
+ب,b
+ت,t
+ث,s
+ج,ɡ
+ح,ħ
+خ,x
+د,d
+ذ,z
+ر,r
+ز,z
+س,s
+ش,ʃ
+ص,sˤ
+ض,dˤ
+ط,tˤ
+ظ,zˤ
+ع,ʕ
+غ,ɣ
+ف,f
+ق,ʔ
+ك,k
+ل,l
+م,m
+ن,n
+ه,h
+و,w
+ي,j
+ء,ʔ
+ئ,ʔ
+ؤ,ʔ
+َ,a
+ِ,i
+ُ,u
+ْ,∅
+ّ,ː
+2,ʔ
+3,ʕ
+5,x
+7,ħ
+8,ɣ
+9,sˤ

epitran_extensions/ast-Latn.csv

@@ -0,0 +1,92 @@
+Orth,Phon
+ieu,jew
+ueu,wew
+ch,tʃ
+ll,ʎ
+rr,r
+qu,k
+gu,ɡ
+gü,ɡw
+ts,ts
+ai,ai
+ei,ei
+oi,oi
+au,aw
+eu,ew
+ou,ow
+ia,ja
+ie,je
+io,jo
+iu,ju
+ua,wa
+ue,we
+ui,wi
+uo,wo
+ñ,ɲ
+Ñ,ɲ
+ḥ,h
+Ḥ,h
+h,
+H,
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+i,i
+j,x
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,ɾ
+s,s
+t,t
+u,u
+v,b
+w,w
+x,ʃ
+y,j
+z,θ
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+I,i
+J,x
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,ɾ
+S,s
+T,t
+U,u
+V,b
+W,w
+X,ʃ
+Y,j
+Z,θ
+á,a
+é,e
+í,i
+ó,o
+ú,u
+Á,a
+É,e
+Í,i
+Ó,o
+Ú,u
+ü,w
+Ü,w

epitran_extensions/azb-Arab.csv

@@ -0,0 +1,39 @@
+Orth,Phon
+او,o
+وو,uː
+ای,i
+پ,p
+چ,tʃ
+ژ,ʒ
+گ,ɡ
+ڭ,ŋ
+آ,ɑː
+ا,ɑ
+ی,j
+ئ,ʔ
+ب,b
+ت,t
+ث,s
+ج,dʒ
+ح,h
+خ,x
+د,d
+ذ,z
+ر,r
+ز,z
+س,s
+ش,ʃ
+ص,s
+ض,z
+ط,t
+ظ,z
+ع,ʔ
+غ,ɣ
+ف,f
+ق,ɡ
+ک,k
+ل,l
+م,m
+ن,n
+و,v
+ه,h

epitran_extensions/bak-Cyrl.csv

@@ -0,0 +1,40 @@
+Orth,Phon
+а,a
+ә,æ
+б,b
+в,v
+г,ɡ
+ғ,ʁ
+д,d
+е,jɛ
+ё,jo
+ж,ʒ
+з,z
+и,i
+й,j
+к,k
+ҡ,q
+л,l
+м,m
+н,n
+ң,ŋ
+о,o
+ө,ø
+п,p
+р,r
+с,s
+һ,h
+т,t
+у,u
+ү,y
+ф,f
+х,x
+ц,ts
+ч,tʃ
+ш,ʃ
+щ,ɕ
+э,ɛ
+ю,ju
+я,ja
+ъ,∅
+ь,∅

epitran_extensions/bam-Latn.csv

@@ -0,0 +1,28 @@
+Orth,Phon
+a,a
+b,b
+c,tʃ
+d,d
+e,ɛ
+ɛ,ɛ
+f,f
+g,ɡ
+h,h
+i,i
+j,ɟ
+k,k
+l,l
+m,m
+n,n
+ɲ,ɲ
+ŋ,ŋ
+o,ɔ
+ɔ,ɔ
+p,p
+r,r
+s,s
+t,t
+u,u
+w,w
+y,j
+z,z

epitran_extensions/ban-Latn.csv

@@ -0,0 +1,26 @@
+Orth,Phon
+ny,ɲ
+ng,ŋ
+a,a
+b,b
+c,tʃ
+d,d
+e,ə
+é,e
+f,f
+g,g
+h,h
+i,i
+j,dʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+r,r
+s,s
+t,t
+u,u
+w,w
+y,j

epitran_extensions/bar-Latn.csv

@@ -0,0 +1,72 @@
+Orth,Phon
+sch,ʃ
+ch,x
+ng,ŋ
+nk,ŋk
+sp,ʃp
+st,ʃt
+pf,pf
+ei,ɔɪ
+au,ɔʊ
+ou,oʊ
+oa,oɐ
+ia,iɐ
+ua,uɐ
+ä,ɛ
+Ä,ɛ
+ö,ø
+Ö,ø
+ü,y
+Ü,y
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,h
+i,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,r
+s,s
+t,t
+u,u
+v,f
+w,v
+x,ks
+y,i
+z,ts
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,h
+I,i
+J,j
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,r
+S,s
+T,t
+U,u
+V,f
+W,v
+X,ks
+Y,i
+Z,ts

epitran_extensions/bel-Cyrl.csv

@@ -0,0 +1,69 @@
+Orth,Phon
+дж,dʒ
+дз,dz
+Дж,dʒ
+Дз,dz
+і,i
+І,i
+ў,w
+Ў,w
+е,jɛ
+ё,jo
+ю,ju
+я,ja
+Е,jɛ
+Ё,jo
+Ю,ju
+Я,ja
+а,a
+б,b
+в,v
+г,ɣ
+д,d
+ж,ʒ
+з,z
+й,j
+к,k
+л,l
+м,m
+н,n
+о,o
+п,p
+р,r
+с,s
+т,t
+у,u
+ф,f
+х,x
+ц,ts
+ч,tʃ
+ш,ʃ
+ы,ɨ
+ь,ʲ
+э,ɛ
+А,a
+Б,b
+В,v
+Г,ɣ
+Д,d
+Ж,ʒ
+З,z
+Й,j
+К,k
+Л,l
+М,m
+Н,n
+О,o
+П,p
+Р,r
+С,s
+Т,t
+У,u
+Ф,f
+Х,x
+Ц,ts
+Ч,tʃ
+Ш,ʃ
+Ы,ɨ
+Ь,ʲ
+Э,ɛ

epitran_extensions/bho-Deva.csv

@@ -0,0 +1,53 @@
+Orth,Phon
+अ,ə
+आ,aː
+इ,i
+ई,iː
+उ,u
+ऊ,uː
+ए,eː
+ऐ,ɛː
+ओ,oː
+औ,ɔː
+क,k
+ख,kʰ
+ग,ɡ
+घ,ɡʱ
+च,tʃ
+छ,tʃʰ
+ज,dʒ
+झ,dʒʱ
+ट,ʈ
+ठ,ʈʰ
+ड,ɖ
+ढ,ɖʱ
+त,t̪
+थ,t̪ʰ
+द,d̪
+ध,d̪ʱ
+न,n
+प,p
+फ,pʰ
+ब,b
+भ,bʱ
+म,m
+य,j
+र,r
+ल,l
+व,ʋ
+श,ʃ
+ष,ʂ
+स,s
+ह,ɦ
+ा,aː
+ि,i
+ी,iː
+ु,u
+ू,uː
+े,eː
+ै,ɛː
+ो,oː
+ौ,ɔː
+ं,̃
+ः,ɦ
+्,

epitran_extensions/bjn-Latn.csv

@@ -0,0 +1,26 @@
+Orth,Phon
+ny,ɲ
+ng,ŋ
+a,a
+b,b
+c,tʃ
+d,d
+e,ə
+é,e
+f,f
+g,g
+h,h
+i,i
+j,dʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+r,r
+s,s
+t,t
+u,u
+w,w
+y,j

epitran_extensions/bod-Tibt.csv

@@ -0,0 +1,35 @@
+Orth,Phon
+ཀ,k
+ཁ,kʰ
+ག,g
+ང,ŋ
+ཅ,tʃ
+ཆ,tʃʰ
+ཇ,dʒ
+ཉ,ɲ
+ཏ,t
+ཐ,tʰ
+ད,d
+ན,n
+པ,p
+ཕ,pʰ
+བ,b
+མ,m
+ཙ,ts
+ཚ,tsʰ
+ཛ,dz
+ཝ,w
+ཞ,ʒ
+ཟ,z
+འ,ʔ
+ཡ,j
+ར,r
+ལ,l
+ཤ,ʃ
+ས,s
+ཧ,h
+ཨ,ʔ
+ི,i
+ུ,u
+ེ,e
+ོ,o

epitran_extensions/bos-Latn.csv

@@ -0,0 +1,31 @@
+Orth,Phon
+dž,dʒ
+lj,ʎ
+nj,ɲ
+a,a
+b,b
+c,ts
+č,tʃ
+ć,tɕ
+d,d
+đ,dʑ
+e,e
+f,f
+g,ɡ
+h,x
+i,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+r,r
+s,s
+š,ʃ
+t,t
+u,u
+v,v
+z,z
+ž,ʒ

epitran_extensions/bpy-Beng.csv

@@ -0,0 +1,50 @@
+Orth,Phon
+ড়,ɽ
+ঢ়,ɽʰ
+য়,j
+ক,k
+খ,kʰ
+গ,g
+ঘ,gʰ
+ঙ,ŋ
+চ,tʃ
+ছ,tʃʰ
+জ,dʒ
+ঝ,dʒʰ
+ঞ,ɲ
+ট,ʈ
+ঠ,ʈʰ
+ড,ɖ
+ঢ,ɖʰ
+ণ,ɳ
+ত,t̪
+থ,t̪ʰ
+দ,d̪
+ধ,d̪ʰ
+ন,n
+প,p
+ফ,pʰ
+ব,b
+ভ,bʰ
+ম,m
+য,dʒ
+র,r
+ল,l
+শ,ʃ
+ষ,ʂ
+স,s
+হ,ɦ
+ৎ,t
+ং,ŋ
+ঃ,h
+্,∅
+া,aː
+ি,i
+ী,iː
+ু,u
+ূ,uː
+ৃ,ri
+ে,e
+ৈ,oi
+ো,o
+ৌ,ou

epitran_extensions/bre-Latn.csv

@@ -0,0 +1,79 @@
+Orth,Phon
+c'h,x
+C'h,x
+ch,ʃ
+Ch,ʃ
+zh,z
+Zh,z
+gn,ɲ
+lh,ʎ
+ou,u
+Ou,u
+ñ,n
+Ñ,n
+à,a
+â,a
+è,ɛ
+é,e
+ê,ɛ
+ë,e
+ì,i
+î,i
+ï,i
+ò,ɔ
+ô,o
+ù,y
+û,y
+ü,y
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,h
+i,i
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,r
+s,s
+t,t
+u,y
+v,v
+w,w
+x,ks
+y,i
+z,z
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,h
+I,i
+J,ʒ
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,r
+S,s
+T,t
+U,y
+V,v
+W,w
+X,ks
+Y,i
+Z,z

epitran_extensions/bug-Latn.csv

@@ -0,0 +1,26 @@
+Orth,Phon
+ng,ŋ
+ny,ɲ
+a,a
+b,b
+c,tʃ
+d,d
+e,ə
+é,e
+g,ɡ
+h,h
+i,i
+j,ɟ
+k,k
+l,l
+m,m
+n,n
+o,ɔ
+p,p
+q,ʔ
+r,r
+s,s
+t,t
+u,u
+w,w
+y,j

epitran_extensions/bul-Cyrl.csv

@@ -0,0 +1,81 @@
+Orth,Phon
+дж,dʒ
+дз,dz
+нк,ŋk
+нг,ŋɡ
+ия,ija
+ие,ije
+ът,ɤt
+та,ta
+то,to
+те,te
+ай,aj
+ей,ej
+ой,oj
+уй,uj
+ьо,jo
+йо,jo
+йе,je
+йа,ja
+йу,ju
+ьа,ja
+а,a
+б,b
+в,v
+г,ɡ
+д,d
+е,ɛ
+ж,ʒ
+з,z
+и,i
+й,j
+к,k
+л,l
+м,m
+н,n
+о,ɔ
+п,p
+р,r
+с,s
+т,t
+у,u
+ф,f
+х,x
+ц,ts
+ч,tʃ
+ш,ʃ
+щ,ʃt
+ъ,ɤ
+ь,j
+ю,ju
+я,ja
+А,a
+Б,b
+В,v
+Г,ɡ
+Д,d
+Е,ɛ
+Ж,ʒ
+З,z
+И,i
+Й,j
+К,k
+Л,l
+М,m
+Н,n
+О,ɔ
+П,p
+Р,r
+С,s
+Т,t
+У,u
+Ф,f
+Х,x
+Ц,ts
+Ч,tʃ
+Ш,ʃ
+Щ,ʃt
+Ъ,ɤ
+Ь,j
+Ю,ju
+Я,ja

epitran_extensions/che-Cyrl.csv

@@ -0,0 +1,119 @@
+Orth,Phon
+кӏ,kʼ
+пӏ,pʼ
+тӏ,tʼ
+цӏ,tsʼ
+чӏ,tʃʼ
+хь,ħ
+хӏ,h
+къ,qʼ
+гӏ,ʕ
+аь,æ
+оь,ø
+уь,y
+юь,jy
+яь,jæ
+дж,dʒ
+дз,dz
+кх,q
+пх,pχ
+тх,tχ
+шт,ʃt
+аа,aː
+ее,eː
+ии,iː
+оо,oː
+уу,uː
+ай,aj
+ей,ej
+ий,ij
+ой,oj
+уй,uj
+ау,aw
+оу,ow
+иу,iw
+ц̣,tsʼ
+ч̣,tʃʼ
+Кӏ,kʼ
+Пӏ,pʼ
+Тӏ,tʼ
+Цӏ,tsʼ
+Чӏ,tʃʼ
+Хь,ħ
+Хӏ,h
+Къ,qʼ
+Гӏ,ʕ
+Аь,æ
+Оь,ø
+Уь,y
+Кх,q
+Дж,dʒ
+Дз,dz
+а,a
+б,b
+в,v
+г,ɡ
+д,d
+е,e
+ё,jo
+ж,ʒ
+з,z
+и,i
+й,j
+к,k
+л,l
+м,m
+н,n
+о,o
+п,p
+р,r
+с,s
+т,t
+у,u
+ф,f
+х,χ
+ц,ts
+ч,tʃ
+ш,ʃ
+щ,ʃtʃ
+ъ,ʔ
+ы,ɨ
+ь,ʲ
+э,e
+ю,ju
+я,ja
+ӏ,ʔ
+А,a
+Б,b
+В,v
+Г,ɡ
+Д,d
+Е,e
+Ё,jo
+Ж,ʒ
+З,z
+И,i
+Й,j
+К,k
+Л,l
+М,m
+Н,n
+О,o
+П,p
+Р,r
+С,s
+Т,t
+У,u
+Ф,f
+Х,χ
+Ц,ts
+Ч,tʃ
+Ш,ʃ
+Щ,ʃtʃ
+Ъ,ʔ
+Ы,ɨ
+Ь,ʲ
+Э,e
+Ю,ju
+Я,ja
+Ӏ,ʔ

epitran_extensions/chv-Cyrl.csv

@@ -0,0 +1,34 @@
+Orth,Phon
+а,a
+ӑ,ə
+б,b
+в,v
+г,ɡ
+д,d
+е,jɛ
+ё,jo
+ж,ʒ
+з,z
+и,i
+й,j
+к,k
+л,l
+м,m
+н,n
+о,o
+п,p
+р,r
+с,s
+ҫ,ɕ
+т,t
+у,u
+ӳ,y
+ф,f
+х,x
+ц,ts
+ч,tʃ
+ш,ʃ
+щ,ɕ
+я,ja
+ъ,∅
+ь,∅

epitran_extensions/cos-Latn.csv

@@ -0,0 +1,64 @@
+Orth,Phon
+ch,k
+gh,ɡ
+sc,sk
+ci,tʃi
+ce,tʃe
+gi,dʒi
+ge,dʒe
+gn,ɲ
+gl,ʎ
+à,a
+è,ɛ
+ì,i
+ò,ɔ
+ù,u
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,
+i,i
+j,j
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,r
+s,s
+t,t
+u,u
+v,v
+z,dz
+À,a
+È,ɛ
+Ì,i
+Ò,ɔ
+Ù,u
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,
+I,i
+J,j
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,r
+S,s
+T,t
+U,u
+V,v
+Z,dz

epitran_extensions/crh-Latn.csv

@@ -0,0 +1,32 @@
+Orth,Phon
+a,a
+b,b
+c,dʒ
+ç,tʃ
+d,d
+e,e
+f,f
+g,g
+ğ,ɣ
+h,h
+ı,ɯ
+i,i
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+ñ,ŋ
+o,o
+ö,ø
+p,p
+q,q
+r,r
+s,s
+ş,ʃ
+t,t
+u,u
+ü,y
+v,v
+y,j
+z,z

epitran_extensions/cym-Latn.csv

@@ -0,0 +1,89 @@
+Orth,Phon
+a,a
+â,aː
+ae,ai
+ai,ai
+au,au
+aw,au
+b,b
+c,k
+ch,χ
+d,d
+dd,ð
+e,ɛ
+ê,eː
+ei,əi
+eu,ɨː
+f,v
+ff,f
+g,ɡ
+h,h
+i,i
+î,iː
+j,dʒ
+l,l
+ll,ɬ
+m,m
+mh,v̥
+n,n
+ng,ŋ
+o,ɔ
+ô,oː
+oe,ɔi
+p,p
+ph,f
+r,r
+rh,r̥
+s,s
+si,ʃ
+t,t
+th,θ
+u,ɨ
+ŵ,uː
+w,w
+y,ə
+ŷ,əː
+A,a
+Â,aː
+Ae,ai
+Ai,ai
+Au,au
+Aw,au
+B,b
+C,k
+Ch,χ
+D,d
+Dd,ð
+E,ɛ
+Ê,eː
+Ei,əi
+Eu,ɨː
+F,v
+Ff,f
+G,ɡ
+H,h
+I,i
+Î,iː
+J,dʒ
+L,l
+Ll,ɬ
+M,m
+Mh,v̥
+N,n
+Ng,ŋ
+O,ɔ
+Ô,oː
+Oe,ɔi
+P,p
+Ph,f
+R,r
+Rh,r̥
+S,s
+Si,ʃ
+T,t
+Th,θ
+U,ɨ
+Ŵ,uː
+W,w
+Y,ə
+Ŷ,əː

epitran_extensions/dan-Latn.csv

@@ -0,0 +1,89 @@
+Orth,Phon
+a,a
+å,ɔ
+æ,ɛ
+ai,ai
+au,au
+b,b
+c,k
+ch,ɕ
+d,d
+dj,dj
+e,e
+eg,ai
+ej,ai
+eu,ɔy
+f,f
+g,ɡ
+gj,j
+h,h
+i,i
+ig,i
+j,j
+k,k
+kj,ɕ
+l,l
+m,m
+n,n
+ng,ŋ
+o,o
+ø,ø
+p,p
+r,ʁ
+rs,ʁs
+s,s
+sj,ɕ
+sk,sg
+skj,ɕ
+t,t
+tj,ɕ
+u,u
+v,v
+w,v
+x,ɡs
+y,y
+z,s
+A,a
+Å,ɔ
+Æ,ɛ
+Ai,ai
+Au,au
+B,b
+C,k
+Ch,ɕ
+D,d
+Dj,dj
+E,e
+Eg,ai
+Ej,ai
+Eu,ɔy
+F,f
+G,ɡ
+Gj,j
+H,h
+I,i
+Ig,i
+J,j
+K,k
+Kj,ɕ
+L,l
+M,m
+N,n
+Ng,ŋ
+O,o
+Ø,ø
+P,p
+R,ʁ
+Rs,ʁs
+S,s
+Sj,ɕ
+Sk,sg
+Skj,ɕ
+T,t
+Tj,ɕ
+U,u
+V,v
+W,v
+X,ɡs
+Y,y
+Z,s

epitran_extensions/diq-Latn.csv

@@ -0,0 +1,33 @@
+Orth,Phon
+a,a
+b,b
+c,dʒ
+ç,tʃ
+d,d
+e,e
+ê,eː
+f,f
+g,g
+h,h
+i,i
+î,iː
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,q
+r,r
+ř,r
+s,s
+ş,ʃ
+t,t
+u,u
+û,uː
+v,v
+w,w
+x,x
+y,j
+z,z

epitran_extensions/ell-Grek.csv

@@ -0,0 +1,127 @@
+Orth,Phon
+ντζ,ndz
+ντσ,nts
+αι,e
+Αι,e
+ΑΙ,e
+ει,i
+Ει,i
+ΕΙ,i
+οι,i
+Οι,i
+ΟΙ,i
+ου,u
+Ου,u
+ΟΥ,u
+ευ,ev
+Ευ,ev
+ΕΥ,ev
+αυ,av
+Αυ,av
+ΑΥ,av
+ηυ,iv
+Ηυ,iv
+ΗΥ,iv
+γγ,ŋɡ
+Γγ,ŋɡ
+ΓΓ,ŋɡ
+γκ,ŋk
+Γκ,ŋk
+ΓΚ,ŋk
+γχ,ŋx
+Γχ,ŋx
+ΓΧ,ŋx
+ντ,nd
+Ντ,nd
+ΝΤ,nd
+μπ,mb
+Μπ,mb
+ΜΠ,mb
+τζ,dz
+Τζ,dz
+ΤΖ,dz
+τσ,ts
+Τσ,ts
+ΤΣ,ts
+σκ,sk
+Σκ,sk
+ΣΚ,sk
+σπ,sp
+Σπ,sp
+ΣΠ,sp
+στ,st
+Στ,st
+ΣΤ,st
+σφ,sf
+Σφ,sf
+ΣΦ,sf
+σχ,sx
+Σχ,sx
+ΣΧ,sx
+αί,e
+εί,i
+οί,i
+ού,u
+ά,a
+έ,e
+ή,i
+ί,i
+ό,o
+ύ,i
+ώ,o
+Ά,a
+Έ,e
+Ή,i
+Ί,i
+Ό,o
+Ύ,i
+Ώ,o
+Α,a
+Β,v
+Γ,ɣ
+Δ,ð
+Ε,e
+Ζ,z
+Η,i
+Θ,θ
+Ι,i
+Κ,k
+Λ,l
+Μ,m
+Ν,n
+Ξ,ks
+Ο,o
+Π,p
+Ρ,r
+Σ,s
+Τ,t
+Υ,i
+Φ,f
+Χ,x
+Ψ,ps
+Ω,o
+α,a
+β,v
+γ,ɣ
+δ,ð
+ε,e
+ζ,z
+η,i
+θ,θ
+ι,i
+κ,k
+λ,l
+μ,m
+ν,n
+ξ,ks
+ο,o
+π,p
+ρ,r
+σ,s
+ς,s
+τ,t
+υ,i
+φ,f
+χ,x
+ψ,ps
+ω,o

epitran_extensions/eus-Latn.csv

@@ -0,0 +1,90 @@
+Orth,Phon
+barren,baren
+iturri,ituri
+berri,beri
+gorri,gori
+zahar,sahar
+zarra,sara
+handi,andi
+txiki,tʃiki
+goiti,goiti
+beiti,beiti
+aurre,aure
+kanpo,kanpo
+mendi,mendi
+zelai,selai
+harri,ari
+landa,landa
+etxe,etʃe
+zuri,suri
+atze,atse
+arte,arte
+gain,gain
+alde,alde
+ibai,ibai
+baso,baso
+erri,eri
+hiri,iri
+tch,tʃ
+uri,uri
+tx,tʃ
+ts,ts
+tz,ts
+dd,ɟ
+tt,c
+rr,r
+ll,ʎ
+ñ,ɲ
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,
+i,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,ɾ
+s,s
+t,t
+u,u
+v,b
+w,w
+x,ʃ
+y,j
+z,s
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,
+I,i
+J,j
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,ɾ
+S,s
+T,t
+U,u
+V,b
+W,w
+X,ʃ
+Y,j
+Z,s
+Ñ,ɲ

epitran_extensions/fao-Latn.csv

@@ -0,0 +1,30 @@
+Orth,Phon
+a,ɛa
+á,ɔa
+b,b
+d,t
+ð,ð
+e,e
+í,ui
+f,f
+g,tʃ
+h,h
+i,i
+j,j
+k,kʰ
+l,l
+m,m
+n,n
+o,ɔ
+ó,ɔu
+p,pʰ
+r,ɹ
+s,s
+t,tʰ
+u,ʉ
+ú,ʉu
+v,v
+y,i
+ý,i
+æ,ai
+ø,ø

epitran_extensions/fij-Latn.csv

@@ -0,0 +1,34 @@
+Orth,Phon
+ngg,ᵑɡ
+dr,ɖr
+mb,ᵐb
+nd,ⁿd
+ng,ŋ
+nj,ɲ
+a,a
+ā,aː
+b,b
+c,tʃ
+d,d
+e,e
+f,f
+g,g
+i,i
+ī,iː
+j,dʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+ō,oː
+p,p
+q,k
+r,r
+s,s
+t,t
+u,u
+ū,uː
+v,v
+w,w
+y,j

epitran_extensions/fil-Latn.csv

@@ -0,0 +1,29 @@
+Orth,Phon
+ng,ŋ
+a,a
+á,a
+b,b
+c,k
+d,d
+e,ɛ
+é,e
+f,f
+g,ɡ
+h,h
+i,i
+í,i
+k,k
+l,l
+m,m
+n,n
+ñ,ɲ
+o,o
+ó,o
+p,p
+r,r
+s,s
+t,t
+u,u
+ú,u
+w,w
+y,j

epitran_extensions/frc-Latn.csv

@@ -0,0 +1,40 @@
+Orth,Phon
+tch,tʃ
+ch,ʃ
+eu,ø
+ou,u
+qu,k
+ss,s
+a,a
+à,a
+â,ɑ
+b,b
+c,k
+d,d
+e,ə
+è,ɛ
+é,e
+ê,ɛ
+f,f
+g,ɡ
+h,
+i,i
+î,i
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+o,ɔ
+ô,o
+p,p
+r,ʁ
+s,s
+t,t
+u,y
+û,y
+v,v
+w,w
+x,ks
+y,j
+z,z

epitran_extensions/frp-Latn.csv

@@ -0,0 +1,43 @@
+Orth,Phon
+tch,tʃ
+ae,ɛ
+ch,ʃ
+eu,ø
+ly,ʎ
+ny,ɲ
+ou,u
+qu,k
+ss,s
+a,a
+à,a
+â,ɑ
+b,b
+c,k
+d,d
+e,ə
+è,ɛ
+é,e
+ê,ɛ
+f,f
+g,ɡ
+h,
+i,i
+î,i
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+o,ɔ
+ô,o
+p,p
+r,ʁ
+s,s
+t,t
+u,y
+û,y
+v,v
+w,w
+x,ks
+y,j
+z,z

epitran_extensions/frr-Latn.csv

@@ -0,0 +1,32 @@
+Orth,Phon
+dj,j
+ng,ŋ
+oo,oː
+a,a
+ä,ɛ
+å,ɔ
+b,b
+d,d
+e,e
+ē,eː
+f,f
+g,g
+h,h
+i,i
+ī,iː
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+ö,ø
+ȫ,øː
+p,p
+r,r
+s,s
+t,t
+u,u
+ü,y
+ū,uː
+w,v

epitran_extensions/fry-Latn.csv

@@ -0,0 +1,93 @@
+Orth,Phon
+a,a
+aa,aː
+ae,ɛː
+ai,ai
+au,au
+b,b
+ch,ɡ
+d,d
+dj,dʒ
+e,ɛ
+ea,ɪə
+ee,eː
+ei,ɛi
+eu,øː
+f,f
+g,ɡ
+h,h
+i,i
+ie,iə
+ij,ɛi
+j,j
+k,k
+l,l
+m,m
+n,n
+ng,ŋ
+o,ɔ
+oa,oə
+oe,u
+oei,ui
+oo,oː
+ou,ɔu
+p,p
+r,r
+s,s
+sj,ʃ
+t,t
+tj,tʃ
+ts,ts
+u,y
+ue,yə
+uu,yː
+v,v
+w,v
+y,i
+z,z
+A,a
+Aa,aː
+Ae,ɛː
+Ai,ai
+Au,au
+B,b
+Ch,ɡ
+D,d
+Dj,dʒ
+E,ɛ
+Ea,ɪə
+Ee,eː
+Ei,ɛi
+Eu,øː
+F,f
+G,ɡ
+H,h
+I,i
+Ie,iə
+Ij,ɛi
+J,j
+K,k
+L,l
+M,m
+N,n
+Ng,ŋ
+O,ɔ
+Oa,oə
+Oe,u
+Oei,ui
+Oo,oː
+Ou,ɔu
+P,p
+R,r
+S,s
+Sj,ʃ
+T,t
+Tj,tʃ
+Ts,ts
+U,y
+Ue,yə
+Uu,yː
+V,v
+W,v
+Y,i
+Z,z

epitran_extensions/fur-Latn.csv

@@ -0,0 +1,41 @@
+Orth,Phon
+ch,k
+gj,ɟ
+gn,ɲ
+ph,f
+qu,kw
+sc,ʃ
+th,t
+a,a
+à,a
+â,a
+b,b
+c,k
+ç,tʃ
+d,d
+e,e
+è,ɛ
+é,e
+f,f
+g,ɡ
+h,
+i,i
+ì,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,ɔ
+ò,ɔ
+p,p
+r,r
+s,s
+t,t
+u,u
+ù,u
+v,v
+w,w
+x,ks
+y,j
+z,ts

epitran_extensions/gla-Latn.csv

@@ -0,0 +1,75 @@
+Orth,Phon
+aoi,ɯː
+bh,v
+ch,x
+dh,ɣ
+fh,f
+gh,ɣ
+mh,v
+ph,f
+sh,h
+th,h
+Bh,v
+Ch,x
+Dh,ɣ
+Fh,f
+Gh,ɣ
+Mh,v
+Ph,f
+Sh,h
+Th,h
+ao,ɯː
+ei,eː
+eu,eː
+ia,iə
+io,iː
+ua,uə
+ui,uː
+èa,ɛː
+òi,ɔː
+à,aː
+è,ɛː
+ì,iː
+ò,ɔː
+ù,uː
+a,a
+b,b
+c,kʰ
+d,t̪
+e,ɛ
+f,f
+g,k
+h,h
+i,i
+l,l
+m,m
+n,n̪
+o,ɔ
+p,pʰ
+r,r
+s,s
+t,tʰ
+u,u
+À,aː
+È,ɛː
+Ì,iː
+Ò,ɔː
+Ù,uː
+A,a
+B,b
+C,kʰ
+D,t̪
+E,ɛ
+F,f
+G,k
+H,h
+I,i
+L,l
+M,m
+N,n̪
+O,ɔ
+P,pʰ
+R,r
+S,s
+T,tʰ
+U,u

epitran_extensions/gle-Latn.csv

@@ -0,0 +1,101 @@
+Orth,Phon
+a,a
+á,aː
+ae,eː
+ai,a
+ao,iː
+b,b
+bh,w
+c,k
+ch,x
+d,d
+dh,ɣ
+e,ɛ
+é,eː
+ea,a
+ei,ɛ
+eo,oː
+f,f
+fh,
+g,ɡ
+gh,ɣ
+h,h
+i,i
+í,iː
+ia,iə
+io,iː
+iu,uː
+j,dʒ
+l,l
+m,m
+mh,w
+n,n
+ng,ŋ
+o,o
+ó,oː
+oi,ɛ
+p,p
+ph,f
+r,ɾ
+s,s
+sh,h
+sl,l̪
+sn,n̪
+t,t̪
+th,h
+u,u
+ú,uː
+ui,i
+v,v
+w,w
+x,ks
+A,a
+Á,aː
+Ae,eː
+Ai,a
+Ao,iː
+B,b
+Bh,w
+C,k
+Ch,x
+D,d
+Dh,ɣ
+E,ɛ
+É,eː
+Ea,a
+Ei,ɛ
+Eo,oː
+F,f
+Fh,
+G,ɡ
+Gh,ɣ
+H,h
+I,i
+Í,iː
+Ia,iə
+Io,iː
+Iu,uː
+J,dʒ
+L,l
+M,m
+Mh,w
+N,n
+Ng,ŋ
+O,o
+Ó,oː
+Oi,ɛ
+P,p
+Ph,f
+R,ɾ
+S,s
+Sh,h
+Sl,l̪
+Sn,n̪
+T,t̪
+Th,h
+U,u
+Ú,uː
+Ui,i
+V,v
+W,w
+X,ks

epitran_extensions/glk-Arab.csv

@@ -0,0 +1,35 @@
+Orth,Phon
+ا,ɑ
+آ,ɑ
+ب,b
+پ,p
+ت,t
+ث,s
+ج,dʒ
+چ,tʃ
+ح,h
+خ,x
+د,d
+ذ,z
+ر,r
+ز,z
+ژ,ʒ
+س,s
+ش,ʃ
+ص,s
+ض,z
+ط,t
+ظ,z
+ع,ʔ
+غ,ɣ
+ف,f
+ق,ɣ
+ک,k
+گ,ɡ
+ل,l
+م,m
+ن,n
+و,u
+ه,h
+ی,i
+ے,e

epitran_extensions/gor-Latn.csv

@@ -0,0 +1,23 @@
+Orth,Phon
+ng,ŋ
+a,a
+b,b
+d,d
+e,e
+f,f
+g,g
+h,h
+i,i
+j,dʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+r,r
+s,s
+t,t
+u,u
+w,w
+y,j

epitran_extensions/gsw-Latn.csv

@@ -0,0 +1,81 @@
+Orth,Phon
+sch,ʃ
+ch,x
+eu,ɔy
+äu,ɔy
+ei,ai
+au,au
+ie,iː
+ue,uː
+öi,øy
+üe,yː
+ä,æ
+ö,ø
+ü,y
+à,a
+è,ɛ
+é,e
+ì,i
+ò,ɔ
+ù,u
+a,a
+b,b
+c,k
+d,d
+e,e
+f,f
+g,ɡ
+h,h
+i,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+q,k
+r,r
+s,s
+t,t
+u,u
+v,f
+w,v
+x,ks
+y,i
+z,ts
+À,a
+È,ɛ
+É,e
+Ì,i
+Ò,ɔ
+Ù,u
+Ä,æ
+Ö,ø
+Ü,y
+A,a
+B,b
+C,k
+D,d
+E,e
+F,f
+G,ɡ
+H,h
+I,i
+J,j
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+Q,k
+R,r
+S,s
+T,t
+U,u
+V,f
+W,v
+X,ks
+Y,i
+Z,ts

epitran_extensions/guj-Gujr.csv

@@ -0,0 +1,47 @@
+Orth,Phon
+ક,k
+ખ,kʰ
+ગ,g
+ઘ,gʰ
+ઙ,ŋ
+ચ,tʃ
+છ,tʃʰ
+જ,dʒ
+ઝ,dʒʰ
+ઞ,ɲ
+ટ,ʈ
+ઠ,ʈʰ
+ડ,ɖ
+ઢ,ɖʰ
+ણ,ɳ
+ત,t̪
+થ,t̪ʰ
+દ,d̪
+ધ,d̪ʰ
+ન,n
+પ,p
+ફ,pʰ
+બ,b
+ભ,bʰ
+મ,m
+ય,j
+ર,r
+લ,l
+વ,ʋ
+શ,ʃ
+ષ,ʂ
+સ,s
+હ,ɦ
+ળ,ɭ
+ા,aː
+િ,i
+ી,iː
+ુ,u
+ૂ,uː
+ે,e
+ૈ,æ
+ો,o
+ૌ,ɔ
+ં,̃
+ઃ,h
+્,∅

epitran_extensions/hat-Latn.csv

@@ -0,0 +1,32 @@
+Orth,Phon
+an,ã
+ch,ʃ
+en,ẽ
+ng,ŋ
+on,õ
+ou,u
+ui,ɥi
+a,a
+b,b
+d,d
+e,e
+è,ɛ
+f,f
+g,g
+h,h
+i,i
+j,ʒ
+k,k
+l,l
+m,m
+n,n
+o,o
+ò,ɔ
+p,p
+r,r
+s,s
+t,t
+v,v
+w,w
+y,j
+z,z

epitran_extensions/hbs-Latn.csv

@@ -0,0 +1,75 @@
+Orth,Phon
+ije,ije
+dž,dʒ
+Dž,dʒ
+DŽ,dʒ
+lj,ʎ
+Lj,ʎ
+LJ,ʎ
+nj,ɲ
+Nj,ɲ
+NJ,ɲ
+je,je
+ie,ie
+č,tʃ
+Č,tʃ
+ć,tɕ
+Ć,tɕ
+đ,dʑ
+Đ,dʑ
+š,ʃ
+Š,ʃ
+ž,ʒ
+Ž,ʒ
+a,a
+b,b
+c,ts
+d,d
+e,e
+f,f
+g,ɡ
+h,x
+i,i
+j,j
+k,k
+l,l
+m,m
+n,n
+o,o
+p,p
+r,r
+s,s
+t,t
+u,u
+v,ʋ
+z,z
+A,a
+B,b
+C,ts
+D,d
+E,e
+F,f
+G,ɡ
+H,x
+I,i
+J,j
+K,k
+L,l
+M,m
+N,n
+O,o
+P,p
+R,r
+S,s
+T,t
+U,u
+V,ʋ
+Z,z
+q,k
+Q,k
+w,v
+W,v
+x,ks
+X,ks
+y,i
+Y,i

epitran_extensions/hsb-Latn.csv

@@ -0,0 +1,37 @@
+Orth,Phon
+dź,dʑ
+ch,x
+a,a
+b,b
+c,ts
+č,tʃ
+ć,tɕ
+d,d
+e,ɛ
+ě,ɪ
+f,f
+g,g
+h,h
+i,i
+j,j
+k,k
+l,l
+ł,w
+m,m
+n,n
+ń,ɲ
+o,ɔ
+ó,o
+p,p
+r,r
+ř,r̝
+s,s
+š,ʃ
+ś,ɕ
+t,t
+u,u
+w,v
+y,ɨ
+z,z
+ž,ʒ
+ź,ʑ

epitran_extensions/hye-Armn.csv

@@ -0,0 +1,97 @@
+Orth,Phon
+յան,jɑn
+յուա,juɑ
+յուե,jue
+յուի,jui
+ույն,uin
+ուտ,ut
+պետ,pɛt
+ուա,uɑ
+ուե,ue
+ուի,ui
+ադր,ɑdɾ
+ագր,ɑɡɾ
+անդ,ɑnd
+արք,ɑɾkʰ
+աստ,ɑst
+աղր,ɑxɾ
+ևր,ɛvɾ
+եզր,ɛzɾ
+երք,ɛɾkʰ
+ութ,ut
+ուղ,uʁ
+ուփ,upʰ
+ուք,ukʰ
+ուֆ,uf
+յար,jɑɾ
+յեկ,jɛk
+յել,jɛl
+յու,ju
+այ,ɑj
+աո,ɑw
+ու,u
+ևի,ɛvi
+եյ,ej
+ոյ,oj
+ավ,ɑv
+ակ,ɑk
+ար,ɑr
+աս,ɑs
+եց,ɛts
+դր,dɾ
+գր,ɡɾ
+խր,xɾ
+չկ,tʃk
+աու,ɑv
+աշ,ɑʃ
+ախ,ɑx
+եբ,ɛb
+և,ɛv
+եդ,ɛd
+եչ,ɛtʃ
+եճ,ɛdʒ
+եզ,ɛz
+եք,ɛkʰ
+են,ɛn
+եպ,ɛp
+եր,ɛɾ
+ես,ɛs
+յև,jɛv
+ա,ɑ
+բ,b
+գ,ɡ
+դ,d
+ե,jɛ
+զ,z
+է,ɛ
+ը,ə
+թ,tʰ
+ժ,ʒ
+ի,i
+լ,l
+խ,x
+ծ,ts
+կ,k
+հ,h
+ձ,dz
+ղ,ʁ
+ճ,tʃ
+մ,m
+յ,j
+ն,n
+շ,ʃ
+ո,o
+չ,tʃʰ
+պ,p
+ջ,dʒ
+ռ,r
+ս,s
+վ,v
+տ,t
+ր,ɾ
+ց,tsʰ
+ւ,v
+փ,pʰ
+ք,kʰ
+օ,o
+ֆ,f

epitran_extensions/hyw-Armn.csv

@@ -0,0 +1,99 @@
+Orth,Phon
+յանթ,jɑntʰ
+յուա,juɑ
+յուե,jue
+յուи,jui
+ուи,ui
+ուտ,ut
+պետ,pɛt
+յան,jɑn
+ուա,uɑ
+ուե,ue
+ուի,ui
+ադր,ɑtʰɾ
+ագր,ɑkʰɾ
+անդ,ɑntʰ
+արք,ɑɾkʰ
+աստ,ɑsd
+ախր,ɑxɾ
+եվր,ɛvɾ
+եզր,ɛzɾ
+երք,ɛɾkʰ
+ութ,utʰ
+ուղ,uʁ
+ուփ,upʰ
+ուք,ukʰ
+ուֆ,uf
+յար,jɑɾ
+յեկ,jɛkʰ
+յել,jɛl
+յու,ju
+այ,ɑj
+աո,ɑw
+ու,u
+ևի,ɛvi
+եյ,ej
+ոյ,oj
+ավ,ɑv
+ակ,ɑkʰ
+ար,ɑr
+աս,ɑs
+եց,ɛtsʰ
+դր,tʰɾ
+գր,kʰɾ
+խր,xɾ
+չկ,tʃʰkʰ
+աւ,ɑv
+աշ,ɑʃ
+ախ,ɑx
+եբ,ɛpʰ
+եվ,ɛv
+եդ,ɛtʰ
+եչ,ɛtʃʰ
+եճ,ɛtʃʰ
+եզ,ɛz
+եք,ɛkʰ
+են,ɛn
+եպ,ɛb
+եր,ɛɾ
+ես,ɛs
+յև,jɛv
+և,ɛv
+ա,ɑ
+բ,pʰ
+գ,kʰ
+դ,tʰ
+ե,jɛ
+զ,z
+է,ɛ
+ը,ə
+թ,tʰ
+ժ,ʒ
+ի,i
+լ,l
+խ,x
+ծ,tsʰ
+կ,kʰ
+հ,h
+ձ,ts
+ղ,ʁ
+ճ,tʃʰ
+մ,m
+յ,j
+ն,n
+շ,ʃ
+ո,o
+չ,tʃʰ
+պ,b
+ջ,dʒ
+ռ,r
+ս,s
+վ,v
+տ,d
+ր,ɾ
+ց,tsʰ
+ւ,v
+փ,pʰ
+ք,kʰ
+օ,o
+ֆ,f