---
title: "Application of the Double Metaphone Algorithm to Amharic"
description: "He describes three informal levels of spelling precision (basic, intermediate, canonical) and a nine-type… 16 references, cited by 3 papers here."
authors: ["Daniel Yacob"]
year: 2004
venue: "arXiv:cs/0408052"
source: "https://arxiv.org/abs/cs/0408052"
doi: "10.48550/arxiv.cs/0408052"
arxiv: "cs/0408052"
licence: "arXiv — no licence stated (read only)"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/yacob-2004-application-double-metaphone-algorithm-amharic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yacob-2004-application-double-metaphone-algorithm-amharic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2004](https://md.research.elamur.ai/free/en/papers/2004/page/2/) / Application of the Double Metaphone Algorithm to Amharic Orthography

# Application of the Double Metaphone Algorithm to Amharic Orthography

**Authors:** Daniel Yacob · **Year:** 2004 · **Venue:** arXiv:cs/0408052 · **Licence:** arXiv — no licence stated (read only)

**Links:** [Publisher's page](https://arxiv.org/abs/cs/0408052) · [PDF](https://arxiv.org/pdf/cs/0408052) · [DOI 10.48550/arxiv.cs/0408052](https://doi.org/10.48550/arxiv.cs/0408052) · [arXiv cs/0408052](https://arxiv.org/abs/cs/0408052) · [OpenAlex](https://openalex.org/W1652017806)

**Citations:** 11 citations (OpenAlex, 2026-10-14) · [3](https://md.research.elamur.ai/free/en/papers/yacob-2004-application-double-metaphone-algorithm-amharic/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper asks whether Metaphone, a phonetic preconditioning step designed for English, can help compare Amharic words for search and spell checking. Yacob argues that the English rules do not carry over to Ethiopic script. He first analyses why Amharic spelling varies. He describes three informal levels of spelling precision (basic, intermediate, canonical) and a nine-type taxonomy of misspellings. These include homophonic symbols inherited from Ge'ez, look-alike glyphs, false Ge'ezisms, assimilations, labiovelar splits, dialect forms, foreign transcriptions and keyboard mistrikes. He then rewrites the Double Metaphone encoding around Amharic orthography and applies it to stemmed, underived words. A test set built from canonical forms and their common variants shows that matching improves as Amharic-specific rules are added. Treating the ወ and የ families like vowels brought the success rate to 96%. The work is exploratory, but it shows that a language-aware phonetic key captures most real spelling variation. It also identifies which error types still need a separate distance algorithm.

## Key points
- The method adapts the Double Metaphone algorithm (Lawrence Philips) to Ethiopic script and is released as the Perl module Text::Metaphone::Amharic on CPAN (v0.09, 2003).
- The paper sorts misspellings into nine types:
  - Syllographic Redundancy
  - Glypheme Misidentification
  - False Ge'ezisms
  - Assimilations and Alternations
  - Orthographic Abbreviations and Elisions
  - Disjoint Labiovelars
  - Dialect Variations
  - Foreign Language Transcription
  - Mistrikes
- Encoding ወ and የ under vowel rules matched 10 more test words and raised the success rate to 96%. Dialect variations (Type 7) that change a syllable's consonant were still hard to match.
- Affixes must be stemmed off before encoding, because the algorithm targets underived words.
- Proposed future work:
  - merge the ጠ/ጸ/ፀ families;
  - give ፐ and ፈ the same code as በ, and perhaps also fold ጀ into ዠ;
  - run a false-positive (over-matching) analysis on a large corpus;
  - build an authoritative Amharic lexicon.

## Interesting ideas
- Normalise words into a coarse phonetic key before measuring edit distance, so that equivalent renderings collapse and only real differences are left for the distance step.
- Build an explicit taxonomy of error sources and evaluate how each category is handled, rather than reporting only one overall score.
- Treat spelling as graded registers (informal, intermediate, canonical) and always offer the canonical form as the target.
- Model errors caused by the input method, such as neighbouring keystrokes or look-alike glyphs, as a separate class of confusions.
- Pair each loosening of a matching rule with a check for over-matching on a large corpus, so that recall gains are not bought with false positives.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (tokens): a Metaphone-style key could map misspelled or variant English words to lexicon entries before lemmatisation, with the original form kept for explainability.
- **I2** (learn): tagging failed tokens or parses by error type, such as typo, dialect spelling or transliteration, would show which repair rules are missing.
- **I3** (tokens): text could be normalised toward canonical spellings while recording the register detected, which later stages could use as a style signal.
- **I4** (tokens): a keyboard-adjacency confusion table could give a cheap, explainable correction candidate for noisy English input.
- **I5** (learn): every new equivalence or repair rule could be tested against negative controls to measure how many wrong merges it introduces.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🎯 Decision Sciences](https://md.research.elamur.ai/free/en/shelves/decision-sciences/) · 209 works · [🎯 Management Science and Operations Research](https://md.research.elamur.ai/free/en/shelves/decision-sciences/management-science-and-operations-research/) · 108 works

🧭 Topics (1): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 2,908 works

## Cite

Cite as: Daniel Yacob. 2004. Application of the Double Metaphone Algorithm to Amharic Orthography. arXiv:cs/0408052.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@misc{yacob2004application,
  title = {{Application of the Double Metaphone Algorithm to Amharic Orthography}},
  author = {Daniel Yacob},
  year = {2004},
  doi = {10.48550/arxiv.cs/0408052},
  eprint = {cs/0408052},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.cs/0408052}
}
```

- [← previous TALP system for the English lexical sample task](https://md.research.elamur.ai/free/en/papers/escudero-2004-talp-system-english-lexical-sample/)
- [next → Network of Data Centres (NetDC): BNSC - An Arabic Broadcast News Speech Corpus](https://md.research.elamur.ai/free/en/papers/choukri-2004-network-data-centres-netdc-bnsc/)
- [↑ back to papers of 2004](https://md.research.elamur.ai/free/en/papers/2004/page/2/)

## References

16 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. AMSALU AKLILU, Aማርኛ-Eንግሊዝኛ መዝገበ ቃላት Amharic-English Dictionary, Kuraz Publishing Agency, 1987 (1979 EC). [find](https://scholar.google.com/scholar?q=AMSALU+AKLILU%2C+A%E1%88%9B%E1%88%AD%E1%8A%9B%2DE%E1%8A%95%E1%8C%8D%E1%88%8A%E1%8B%9D%E1%8A%9B+%E1%88%98%E1%8B%9D%E1%8C%88%E1%89%A0+%E1%89%83%E1%88%8B%E1%89%B5+Amharic%2DEnglish+Dictionary%2C+Kuraz+Publishing+Agency%2C+1987+%281979+EC%29%2E)
2. DANIEL YACOB, Text::Metaphone::Amharic, [http://search.cpan.org/author/DYACOB/Text-Metaphone-Amharic/](http://search.cpan.org/author/DYACOB/Text-Metaphone-Amharic/), CPAN, v0.09, 2003. [↗](http://search.cpan.org/author/DYACOB/Text-Metaphone-Amharic/)
3. ETHIOPIAN BIBLE SOCIETY, መጽሐፍ፡ቅዱስ። የብሉይና፡የሐዲስ፡ኪዳን፡መጻሕፍት (mäshäf qədus yäbluyna ḥ ädis kidan mäsahfət), United Bible Societies, Nairobi, 1998. [find](https://scholar.google.com/scholar?q=ETHIOPIAN+BIBLE+SOCIETY%2C+%E1%88%98%E1%8C%BD%E1%88%90%E1%8D%8D%E1%8D%A1%E1%89%85%E1%8B%B1%E1%88%B5%E1%8D%A2+%E1%8B%A8%E1%89%A5%E1%88%89%E1%8B%AD%E1%8A%93%E1%8D%A1%E1%8B%A8%E1%88%90%E1%8B%B2%E1%88%B5%E1%8D%A1%E1%8A%AA%E1%8B%B3%E1%8A%95%E1%8D%A1%E1%88%98%E1%8C%BB%E1%88%95%E1%8D%8D%E1%89%B5+%28m%C3%A4sh%C3%A4f+q%C9%99dus+y%C3%A4bluyna+h%CC%A3+%C3%A4dis+kidan+m%C3%A4sahf%C9%99t%29%2C+United+Bible+Societies%2C+Nairobi%2C+1998%2E)
4. ETHIOPIAN NEWS HEADLINES, Newspaper Archives, [http://archives.news.com.et/](http://archives.news.com.et/), Addis Ababa, 1989-1995 EC. [↗](http://archives.news.com.et/)
5. FISSEHATSION KASSA, መርኆ Aዳምና ግEዝ (märho ʾ ädamna g ʿ əz) , Washington, DC, 1991 (1983 EC). [find](https://scholar.google.com/scholar?q=FISSEHATSION+KASSA%2C+%E1%88%98%E1%88%AD%E1%8A%86+A%E1%8B%B3%E1%88%9D%E1%8A%93+%E1%8C%8DE%E1%8B%9D+%28m%C3%A4rho+%CA%BE+%C3%A4damna+g+%CA%BF+%C9%99z%29+%2C+Washington%2C+DC%2C+1991+%281983+EC%29%2E)
6. HIOB LUDOLF, Lexicon Amharico-Latinum, Frankfurt am Main (Zunner), 1698. [find](https://scholar.google.com/scholar?q=HIOB+LUDOLF%2C+Lexicon+Amharico%2DLatinum%2C+Frankfurt+am+Main+%28Zunner%29%2C+1698%2E)
7. KIDANE WOLD KIFLE, መጽሐፈ፡ሰዋስው፡ወግስ፡ወመዝገበ፡ቃላት፡ሐዲስ (mäshäfä säwasəẉ wägəs wämäzgäb qalat ḥ ädis), Artistic Printers, 1955 (1948 EC). [find](https://scholar.google.com/scholar?q=KIDANE+WOLD+KIFLE%2C+%E1%88%98%E1%8C%BD%E1%88%90%E1%8D%88%E1%8D%A1%E1%88%B0%E1%8B%8B%E1%88%B5%E1%8B%8D%E1%8D%A1%E1%8B%88%E1%8C%8D%E1%88%B5%E1%8D%A1%E1%8B%88%E1%88%98%E1%8B%9D%E1%8C%88%E1%89%A0%E1%8D%A1%E1%89%83%E1%88%8B%E1%89%B5%E1%8D%A1%E1%88%90%E1%8B%B2%E1%88%B5+%28m%C3%A4sh%C3%A4f%C3%A4+s%C3%A4was%C9%99w%CC%A3+w%C3%A4g%C9%99s+w%C3%A4m%C3%A4zg%C3%A4b+qalat+h%CC%A3+%C3%A4dis%29%2C+Artistic+Printers%2C+1955+%281948+EC%29%2E)
8. LAWRENCE PHILIPS, The Double Metaphone Search Algorithm, C/C++ Users Journal, June 2000. [find](https://scholar.google.com/scholar?q=LAWRENCE+PHILIPS%2C+The+Double+Metaphone+Search+Algorithm%2C+C%2FC%2B%2B+Users+Journal%2C+June+2000%2E)
9. MAURICE AUBREY, Text::Double Metaphone, [http://search.cpan.org/author/MAURICE/Text-Double](http://search.cpan.org/author/MAURICE/Text-Double) Metaphone/, CPAN, v0.05, 2001. [↗](http://search.cpan.org/author/MAURICE/Text-Double)
10. NARAYANA K. MURTHY, Spell Checkers, University of Hyderabad, Hyderabad, 2002. [find](https://scholar.google.com/scholar?q=NARAYANA+K%2E+MURTHY%2C+Spell+Checkers%2C+University+of+Hyderabad%2C+Hyderabad%2C+2002%2E)
11. ROGER COWLEY, “The Standardization of Amharic Spelling”, in Journal of Ethiopian Studies, 5:2 (1-18), 1967. [find](https://scholar.google.com/scholar?q=ROGER+COWLEY%2C+%E2%80%9CThe+Standardization+of+Amharic+Spelling%E2%80%9D%2C+in+Journal+of+Ethiopian+Studies%2C+5%3A2+%281%2D18%29%2C+1967%2E)
12. ROGER MITTON, Spellchecking by Computer, Journaal of the Simplified Spelling Society, J20, 1996. [find](https://scholar.google.com/scholar?q=ROGER+MITTON%2C+Spellchecking+by+Computer%2C+Journaal+of+the+Simplified+Spelling+Society%2C+J20%2C+1996%2E)
13. SHERRY KLINE, Soundex/Miracode, [http://www.rootsweb.com/~ksscgs/soundex.htm](http://www.rootsweb.com/~ksscgs/soundex.htm), Sumner County Genealogy Society, 2001. [↗](http://www.rootsweb.com/~ksscgs/soundex.htm)
14. TESEMA HABTE MIKAEL GISEW, የAማርኛ መዝገበ ቃላት (yä ʾ ämärñä mäzgäbä qalat), Addis Ababa, 1958 (1951 EC). [find](https://scholar.google.com/scholar?q=TESEMA+HABTE+MIKAEL+GISEW%2C+%E1%8B%A8A%E1%88%9B%E1%88%AD%E1%8A%9B+%E1%88%98%E1%8B%9D%E1%8C%88%E1%89%A0+%E1%89%83%E1%88%8B%E1%89%B5+%28y%C3%A4+%CA%BE+%C3%A4m%C3%A4r%C3%B1%C3%A4+m%C3%A4zg%C3%A4b%C3%A4+qalat%29%2C+Addis+Ababa%2C+1958+%281951+EC%29%2E)
15. WOLF LESLAU, A Reference Grammar Amharic, Otto Harrassowitz Verlag, Wiesbaden, Germany, 1995. [find](https://scholar.google.com/scholar?q=WOLF+LESLAU%2C+A+Reference+Grammar+Amharic%2C+Otto+Harrassowitz+Verlag%2C+Wiesbaden%2C+Germany%2C+1995%2E)
16. YESHIMEBET T BELAY, The Ethiopian Yellow Pages 8 th Edition, Feker Inc, Alexandria, VA, 2002. [find](https://scholar.google.com/scholar?q=YESHIMEBET+T+BELAY%2C+The+Ethiopian+Yellow+Pages+8+th+Edition%2C+Feker+Inc%2C+Alexandria%2C+VA%2C+2002%2E)

*Updated 29 October 2026*