Qwen3 Voice Embedding Collection Standalone ECAPA-TDNN x-vector speaker encoders extracted from Qwen3-TTS. 1024-dim (0.6B) and 2048-dim (1.7B). • 4 items • Updated 1 day ago • 20
view article Article GGML and llama.cpp join HF to ensure the long-term progress of Local AI +4 5 days ago • 376