SKILL-KD: Contrastive Skill Distillation for LLM Agents Paper • 2607.28048 • Published 11 days ago • 14
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 Feb 27 • 29
Running Agents Featured 103 MOSS Transcribe Diarize 🏢 103 Transcribe audio/video with speaker diarization
nvidia/diar_streaming_sortformer_4spk-v2 Automatic Speech Recognition • 0.1B • Updated 2 days ago • 47.3k • 135
Running on CPU Upgrade Featured 3.27k The Smol Training Playbook 📚 3.27k The secrets to building world-class LLMs