Fix standalone model loading code - verified to work 82e6778 verified HuaminChen commited on about 15 hours ago
Make README fully standalone - no external repo needed e737256 verified HuaminChen commited on about 15 hours ago
Add standalone loading option using only transformers 6445452 verified HuaminChen commited on about 15 hours ago
Fix usage instructions with proper installation steps 56f3801 verified HuaminChen commited on about 15 hours ago
Fix total parameter count: ~120M not ~250M 4817d62 verified HuaminChen commited on about 15 hours ago
Document both model.pt and model.safetensors formats 818f7d0 verified HuaminChen commited on about 15 hours ago
Remove redundant pytorch_model.bin (use model.pt or model.safetensors) ec915da verified HuaminChen commited on about 15 hours ago
Update model.safetensors with Stage 5 audio-aligned weights 4485d23 verified HuaminChen commited on about 15 hours ago
Fix loading example to match checkpoint format 75eae3f verified HuaminChen commited on about 15 hours ago
Fix Whisper-tiny encoder param count (8M, not 39M) b4136e7 verified HuaminChen commited on about 15 hours ago
Update model with audio-text alignment (Stage 5: R@1=36.38%) e6b2249 verified HuaminChen commited on about 15 hours ago
HuaminChen
commited on
