cstr/confucius4-tts-GGUF
Add encoder-only w2v-BERT 2.0 (17L) GGUF for native speaker conditioning
Add model card (Apache-2.0, attribution)
Add S2A Q4_K GGUF (Confucius4-TTS)
Add S2A Q8_0 GGUF (Confucius4-TTS)
Add S2A F16 GGUF (Confucius4-TTS, Flow-matching DiT+WaveNet)
Add T2S Q4_K GGUF (Confucius4-TTS)
Add T2S Q8_0 GGUF (Confucius4-TTS)
Add T2S F16 GGUF (Confucius4-TTS, GPT-2 24L/1280d/20h)
Add T2S F16 GGUF (Confucius4-TTS, GPT-2 24L/1280d/20h)
Add BigVGAN v2 22kHz Q8_0 GGUF
Add BigVGAN v2 22kHz F16 GGUF (80-mel vocoder for Confucius4-TTS)
Add tokenizer.ggml.tokens + scores to T2S F16 GGUF
Add T2S Q4_K GGUF (Confucius4-TTS)
Add T2S Q8_0 GGUF (Confucius4-TTS)
Add T2S F16 GGUF (Confucius4-TTS, GPT-2 24L/1280d/20h)
Add S2A Q4_K GGUF (Confucius4-TTS)
Add S2A Q8_0 GGUF (Confucius4-TTS)
Add S2A F16 GGUF (Confucius4-TTS, Flow-matching DiT+WaveNet)
Add T2S Q4_K GGUF (Confucius4-TTS)
Add T2S Q8_0 GGUF (Confucius4-TTS)
Add T2S F16 GGUF (Confucius4-TTS, GPT-2 24L/1280d/20h)
Add T2S F16 GGUF (Confucius4-TTS, GPT-2 24L/1280d/20h)
initial commit