← All examples

ModelEthio-ASR Amharic (600M)10Ethio-ASR Multi (300M)8Ethio-ASR Multi (600M)7Ethio-ASR Multi (94M)4Meta MMS-1B19OmniASR-CTC (1B)2OmniASR-CTC (300M)15OmniASR-CTC (3B)0OmniASR-LLM (1B)0OmniASR-LLM (300M)0OmniASR-LLM (3B)0OpenAI Whisper-small220
Audio qualityClean Studio19Ambient Room Noise19Harsh Environment19Distant / Low-Volume Mic19Wideband Phone (VoLTE)19Randomized Phone Channel19Narrowband 2G + Packet Loss19

What did the model get wrong?

Meta MMS-1B Clean Studio   Example 1 of 20  ·  speaker: Female

🎧 Listen — Clean Studio
Clear, studio-quality recording. Best-case audio.
✅ What was actually said
ሲ ሲ ቲ ሁለት ዜሮ ሁለት አምስት አንድ ሁለት አንድ ሰባት ሁለት አራት ሰባት ስምንት ዜሮ ሰባት ዘጠኝ አራት
🤖 What this model heard
ይ ይበ ል ል ር ር ወ ር ት
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
9
Wrong
10
Missed
0
Extra
0%
Words right
Out of 19 spoken words, this model got 0 right (0%). Errors: 9 words wrong; 10 missed.

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