Boson AI HuBERT Base A general purpose HuBERT Base checkpoint released by Boson AI, used inside the Higgs Audio Tokenizer as the semantic teacher. What it is Standard HuBERT Base architecture (12 transformer layers, hidden size 768, ~95M params) 16 kHz audio input Loadable via AutoModel with trust remote code=True Outputs 768 dim per layer hidden states ( output hidden states=True ) How it is used in Higgs Audio The Higgs Audio Tokenizer distills semantic features from this HuBERT into its semantic branch. From boson multimodal/audio processing/higgs audio tokenizer.py ( semantic techer="hubert base general" ): Direct usage License Apache 2.0.
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy