← All examples

ModelEthio-ASR Amharic (600M)9Ethio-ASR Multi (300M)13Ethio-ASR Multi (600M)10Ethio-ASR Multi (94M)9Meta MMS-1B12OmniASR-CTC (1B)6OmniASR-CTC (300M)8OmniASR-CTC (3B)4OmniASR-LLM (1B)1OmniASR-LLM (300M)9OmniASR-LLM (3B)4OpenAI Whisper-small12
Audio qualityClean Studio0Ambient Room Noise1Harsh Environment2Distant / Low-Volume Mic0Wideband Phone (VoLTE)0Randomized Phone Channel0Narrowband 2G + Packet Loss9

What did the model get wrong?

OmniASR-LLM (300M) Narrowband 2G + Packet Loss   Example 9 of 20  Âˇ  speaker: Female

🎧 Listen — Narrowband 2G + Packet Loss
Heavy background noise plus phone-call compression — worst-case stress test.
✅ What was actually said
ቁጥሩ ዜሮ ዘጠኝ አንድ ሁለት አምስት ስድስት ሰባት ስምንት ዘጠኝ ዜሮ ነው
🤖 What this model heard
ቁጥሩ የሮበተኛ አምስት ሁኔታ አምስት ስደት አዋት ስነ እንጀራ የእንጀራ ነው
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
3
Correct
8
Wrong
1
Missed
0
Extra
25%
Words right
Out of 12 spoken words, this model got 3 right (25%). Errors: 8 words wrong; 1 missed.

Try switching Audio quality — see the same model degrade as the audio gets noisier.