> ## Documentation Index
> Fetch the complete documentation index at: https://afri-health-ai.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Supported Languages and Code-Switching Roadmap

> See which Ethiopian language pairs AfriHealth AI supports today, how code-switched speech is handled, and when Afaan Oromoo and Tigrinya are planned.

AfriHealth AI is designed for code-switched speech, where clinicians and patients mix a local language with English in the same sentence. Speech is transcribed by the Intron Sahara v2.5 ASR engine, which keeps local disease descriptions grounded without translating them.

## Language scope

| Language pair | Status |
| - | - |
| Amharic-English | Active (v1.0) |
| Afaan Oromoo-English | Roadmap Phase 2 |
| Tigrinya-English | Roadmap Phase 3 |

## Example of code-switched input

From Gold Standard Case 3 (Pediatric Respiratory & High Fever):

> ልጅዎን ምን አመመው? severe cough እና high fever...

The transcript keeps both scripts as spoken, so clinicians can review the exact wording.

<Tip>
  Select your language scope in the Frontline Triage module before recording. Only Amharic-English is active in v1.0.
</Tip>

## Related pages

* [Frontline Triage](/modules/frontline-triage)
* [Benchmarks](/evidence/benchmarks)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.