Addis AI

Translation

High-accuracy neural machine translation for African languages.

The Translation API leverages our specialized, fine-tuned models designed to master the linguistic complexity of Amharic (am), Afan Oromo (om), and English (en).

Our engine is rigorously trained on real-world conversational data and culturally accurate datasets. This ensures exceptional grammatical precision and deep cultural relevance, delivering translations that feel natural and authentic across these three key languages.

Use Cases

Public Service

Automatically translate government forms, announcements, and news between Amharic and Afan Oromo to reach wider audiences.

Marketplace

Enable real-time chat translation between buyers and sellers who speak different local languages or English.

Education

Translate educational materials and technical documentation from English into local languages to improve accessibility.


Usage Guide

Basic Translation

Translate text between any of the supported languages (am, om, en).

import AddisAI from "addisai";

const addis = new AddisAI();
const result = await addis.translate.create({
  text: "ገበሬው ማሳውን ዘወትር ይጎበኛል",
  from: "am",
  to: "om",
});

console.log(result.text);
from addisai import AddisAI

addis = AddisAI()
result = addis.translate.create(
    text="ገበሬው ማሳውን ዘወትር ይጎበኛል",
    source="am",
    target="om",
)

print(result["text"])
curl --location 'https://api.addisassistant.com/api/v1/translate' \\
  --header 'Content-Type: application/json' \\
  --header 'x-api-key: $ADDIS_API_KEY' \\
  --data '{
    "text": "ገበሬው ማሳውን ዘወትር ይጎበኛል",
    "source_language": "am",
    "target_language": "om"
  }'

Available Parameters

ValueNode.jsPythonRESTRequired
Text to translatetexttexttextYes
Source languagefromsourcesource_languageYes: am, om, or en
Target languagetotargettarget_languageYes: am, om, or en

Source and target must be different. Translation is supported in both directions between Amharic, Afan Oromo, and English.


API Reference

Request Parameters

Prop

Type

Response Schema

{
    "status": "success",
    "data": {
        "translation": "Qotee bulaan yeroo mara maasaa isaa ni daawwata.",
        "source_language": "am",
        "target_language": "om",
        "quality": "high",
        "usage_metadata": {
            "prompt_token_count": 113,
            "candidates_token_count": 57,
            "total_token_count": 170
        }
    }
}

Prop

Type


Best Practices

Ensure the highest quality translations with these guidelines.

Input Fidelity

Spelling Matters: Ensure the source text has correct spelling. For Amharic, consistency in Fidel usage (e.g., using vs correctly) improves output quality.

Punctuation: Include proper punctuation (, ?) to help the model understand sentence boundaries.

Supported Languages

Triangular Support: We support full bidirectional translation between all three languages:

Amharic (am)Oromo (om)English (en)

Context

Full Sentences: Avoid translating single words in isolation if possible. Context helps resolve ambiguities (e.g., "eye" vs "spring" in Amharic).

On this page