Ultra-Low Bit Post-Training Quantization of Large Speech Models via K-Means Clustering and Mixed Precision Allocation

Tianteng Gu, Bei Liu 0003, Haoyu Wang 0007, Yanmin Qian. Ultra-Low Bit Post-Training Quantization of Large Speech Models via K-Means Clustering and Mixed Precision Allocation. In Odette Scharenborg, Catharine Oertel, Khiet Truong, editors, 26th Annual Conference of the International Speech Communication Association, Interspeech 2025, Rotterdam, The Netherlands, 17-21 August 2025. ISCA, 2025. [doi]

Abstract

Abstract is missing.