[ICML2026] SPEAR: A Unified SSL Framework for Learning Speech and Audio Representations
Xiaoyu Yang
marcoyang
AI & ML interests
ASR, Machine learning
Recent Activity
published a dataset about 8 hours ago
marcoyang/easy_english_dataset updated a dataset about 8 hours ago
marcoyang/easy_english_dataset updated a model 25 days ago
marcoyang/SALMONN-2-8BOrganizations
models 51
marcoyang/SALMONN-2-8B
Feature Extraction • 9B • Updated • 1.08k • 3
marcoyang/spear-xlarge-speech-audio
Feature Extraction • 0.6B • Updated • 2.18k • 7
marcoyang/spear-xlarge-speech-audio-v2
Feature Extraction • 0.6B • Updated • 1.69k • 8
marcoyang/spear-xlarge-speech-audio-v2-icefall
Updated
marcoyang/spear-base-speech-audio-v2
93.3M • Updated • 1.48k
marcoyang/spear-base-speech-v2
93.3M • Updated • 44
marcoyang/spear-base-speech-audio
93.3M • Updated • 12 • 2
marcoyang/spear-base-speech
93.3M • Updated • 118
marcoyang/spear-large-speech-audio
0.3B • Updated • 319
marcoyang/spear-large-speech
0.3B • Updated • 47 • 1