Kai Yu

Real-world conversational dynamics significantly challenge target speaker extraction, revealing that even advanced systems struggle with natural overlap and noise.

MBD-LMs achieve a 2x increase in decoding throughput while maintaining high accuracy, revolutionizing the efficiency of diffusion-based text generation.

Finally, a speech tokenizer that doesn't require extra optimization tricks to work robustly for both generation and understanding tasks in a unified architecture.

Identity-preserving video generation just got a whole lot more faithful: FaithfulFaces maintains identity even under extreme pose variations and occlusions, a feat previous methods struggled with.

Adversarial training doesn't have to hurt speaker verification: by explicitly modeling language, you can disentangle speaker and language characteristics without sacrificing speaker discriminability.

Forget expensive audio-text data collection: TASU2 lets you dial in the perfect amount of noise for training your speech LLM, all from text.

G-STAR tackles long-form, multi-speaker ASR by giving Speech-LLMs time-aware speaker tracking, enabling robust identity linking across chunks.