โ† All examples

ModelEthio-ASR Amharic (600M)5Ethio-ASR Multi (300M)9Ethio-ASR Multi (600M)8Ethio-ASR Multi (94M)4Meta MMS-1B12OmniASR-CTC (1B)0OmniASR-CTC (300M)2OmniASR-CTC (3B)0OmniASR-LLM (1B)0OmniASR-LLM (300M)0OmniASR-LLM (3B)0OpenAI Whisper-small9
Audio qualityClean Studio12Ambient Room Noise12Harsh Environment12Distant / Low-Volume Mic12Wideband Phone (VoLTE)12Randomized Phone Channel12Narrowband 2G + Packet Loss12

What did the model get wrong?

Meta MMS-1B Wideband Phone (VoLTE)   Example 20 of 20  ยท  speaker: Male

๐ŸŽง Listen โ€” Wideband Phone (VoLTE)
Compressed, narrow-band audio like a regular phone call.
โœ… What was actually said
แ‰แŒฅแˆฉ แ‹œแˆฎ แ‹˜แŒ แŠ แˆแˆˆแ‰ต แ‹œแˆฎ แŠ แˆแˆตแ‰ต แˆตแ‹ตแˆตแ‰ต แˆฐแ‰ฃแ‰ต แˆตแˆแŠ•แ‰ต แ‹˜แŒ แŠ แ‹œแˆฎ แАแ‹
๐Ÿค– What this model heard
แ‰ณแ‰ แˆ˜ แ‹ˆ แ‹ˆ
Correct
Wrong (different word)
Missed (skipped)
Extra (added)
0
Correct
3
Wrong
9
Missed
0
Extra
0%
Words right
Out of 12 spoken words, this model got 0 right (0%). Errors: 3 words wrong; 9 missed.

Try switching Audio quality โ€” see the same model degrade as the audio gets noisier.