Amharic Speech Research

Live ASR Lab

Loading model…
Start microphone
Speak naturally in Amharic. Partials update during speech; a pause finalizes the utterance.
Live hypothesisIdle
Your partial transcription will appear here.
Finalized utterances

Review as you speak

AcceptCorrectReject
No finalized utterances yet.
Rights-aware corpus QA

Podcast review queue

Only sources with explicit reuse permission can be marked training-eligible. Other public podcasts can be benchmarked separately but stay out of the training export.
Reviewed0
Reviewed WER
Reviewed CER
Loading podcast catalog…
What is live here?

Incremental wrapper, not a causal encoder

The current 606M Phase37 w2v-BERT checkpoint scores 21.84% WER on the fixed development benchmark. The heavier offline research stack reached 18.63% but is not used for interactive partials. This wrapper produces partial text every ~4 seconds and finalizes on speech pauses or a 12-second cap.

Why this matters

Collect real failure cases

Each final utterance is stored with audio, model fingerprint, latency and your QA correction. That gives us high-value supervised examples for later ASR improvement.

Later speech generation

ASR-approved ≠ TTS-approved

Podcast and live audio can help recognition, but TTS requires additional speaker consent, identity, quality and voice-use rights. Those fields remain separate.