Recently updated channels

Pick one to see only that channelBrowse all channels →
Matthew Bermantoday +316AI Engineertoday +6Theo - t3․ggtodayThe Builderstoday +1a16ztoday +2硅谷101today +1SemiAnalysistoday +1Chris WilliamsontodayCheeky PinttodayInvest Like The Besttoday

Results

1 articles
Sort
TopicVoice AILanguageSimplified Chinese
16 slides23:39ZH → English
跟李沐学AI· Jul 23, 2025Li Mu: Adding Ten Million Hours of Speech to a Text LLM, Open-Sourcing Higgs Audio李沐 · Boson AILi Mu explains how the team added a huge amount of speech data to text LLM training, built an AudioLLM that can listen and speak, and open-sourced Higgs Audio. He covers architecture trade-offs, token compression, where the data came from, and how to run it on a 3090.Voice AIMultimodal AIModels & training+2