Model · Phos
A speech-to-text model with translation built for African languages.
Phos is designed to listen, transcribe, and translate in ways that are useful for real communities — from customer support to education and public services. It reads spoken Krio and produces text you can act on.
Core capability
Most speech pipelines treat transcription and translation as two separate, lossy steps. Phos does both in a single workflow, so spoken language in a low-resource African language is transcribed and meaningfully converted into another language at once.
Model details
- Task
- Speech-to-text with translation
- Languages
- Krio, English
- Access
- Hosted API / playground
- First released
- August 2026
- Evaluation (WER, chrF, latency)
- Pending publication
Evaluation methodology and results will be published here and in the research section as the benchmark set is finalised.
What it can do
Speech-to-text transcription for low-resource African languages
Translation across Krio and English
Real-time conversational workflows for accessibility and service delivery
Multilingual understanding for education, healthcare, and public services