Phone Recognition on PRiSM Multilingual Datasets
16.8PFER (DRC)ZIPA-CTC-NS
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ZIPA-CTC-NSModel Category=Multilingual PR Models2026.03 | 16.8 | 17.1 | 23.1 | 19 | |
| PhoneticXEUSModel Category=Multilingual PR Models2026.03 | 16.8 | 14.4 | 21.9 | 17.7 | |
| POWSMModel Category=Multilingual PR Models2026.03 | 17.1 | 17.1 | 22 | 18.7 | |
| W2V2P-XLSR53Model Category=Multilingual PR Models2026.03 | 17.3 | 13.9 | 31.9 | 21 | |
| W2V2P-LV60Model Category=Multilingual PR Models2026.03 | 17.8 | 15.7 | 24.9 | 19.5 | |
| ZIPA-CTCModel Category=Multilingual PR Models2026.03 | 18 | 17 | 23.7 | 19.6 | |
| POWSM-CTCModel Category=Multilingual PR Models2026.03 | 18.1 | 15.3 | 32.2 | 21.9 | |
| MultiIPAModel Category=Multilingual PR Models2026.03 | 18.3 | 15.2 | 30.5 | 21.3 | |
| KoelLabs-XLSRModel Category=English-centric PR Models2026.03 | 22.9 | 18 | 24.7 | 21.9 | |
| HuPER-RecognizerModel Category=English-centric PR Models2026.03 | 32 | 26.5 | 29.8 | 28.2 | |
| Gemini 2.5 FlashModel Category=Large Audio Language Models, Uncertainty about training data=true2026.03 | 105.3 | 19.7 | 36.3 | 53.8 | |
| Qwen3-Omni-InstructModel Category=Large Audio Language Models, Uncertainty about training data=true2026.03 | 150.2 | 49 | 117.1 | 105.4 |