Yesterday
-
README: explain the llama.cpp + codec.cpp split; llama.rn usage
-
README: user-facing model card (files, run with codec.cpp tts-cli, license as upstream)
3 days ago
-
README: precision notes and changelog
-
Regenerate DAC.speech Q8_0/Q5_K_M/Q4_K_M with quantizable conv layouts (needs codec.cpp 961162a+)
-
Regenerate DAC.speech Q8_0/Q5_K_M/Q4_K_M with quantizable conv layouts (needs codec.cpp 961162a+)
-
Regenerate DAC.speech Q8_0/Q5_K_M/Q4_K_M with quantizable conv layouts (needs codec.cpp 961162a+)
-
Fix NeMo Q5 GGUF
-
Fix NeMo Q8 GGUF
-
Fix WavTokenizer Q8 GGUF
-
Fix WavTokenizer Q4 GGUF
-
Fix WavTokenizer Q5 GGUF
-
Fix remaining codec quantized GGUF files
-
Fix DAC speech GGUF conversions
-
Refresh codec GGUF exports from latest codec.cpp
-
Add refreshed codec.cpp GGUF outputs for DAC, WavTokenizer, and NeMo Nano Codec
-
Delete wavtokenizer-Q4_K_M.gguf
-
Upload folder using huggingface_hub
-
Delete wavtokenizer-Q5_K_M.gguf
-
Delete wavtokenizer.gguf
-
Upload 2 files
-
Upload 3 files
-
Upload wavtokenizer.gguf
-
initial commit
