Automatic Speech Recognition on DefinedAI In-domain (test)
7.5WERMulti-turn ASR (Raw Context)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Multi-turn ASR (Raw Context)Configuration=Multi-turn, Raw, Ntrain=R[1,10], Ndecode=52026.03 | 7.5 | 13.3 | |
| Multi-turn ASR (Raw Context)Configuration=Multi-turn, Raw, Ntrain=R[1,10], Ndecode=102026.03 | 7.5 | 13.1 | |
| PHI-4-MULTIMODAL (Fine-tuned)Configuration=Single-turn, Ntrain=0, Ndecode=02026.03 | 7.6 | 13.5 | |
| Abstract CompressionConfiguration=Multi-turn, Compressed, Ntrain=0→10, Ndecode=102026.03 | 8 | 13.3 | |
| PHI-4-MULTIMODAL (Fine-tuned)Configuration=Single-turn, Ntrain=0, Ndecode=52026.03 | 9.1 | 15.1 | |
| PHI-4-MULTIMODAL (No Fine-tuning)Configuration=Single-turn†, Ntrain=0, Ndecode=02026.03 | 13.4 | 21 |