Unverified50% confidenceFactExact time
Google早期发布的Micro Speech示例模型仅约18KB，能在微控制器上实时识别yes和no等简单指令
1
Sources
50%
Confidence
Long-term
Relevance
8/5/2026
First Seen
Sources
Related Claims
Unverified谷歌的Universal Speech Model与Meta的SeamlessStreaming系统采用了单调注意力（Monotonic Attention）机制来约束解码器按时序推进65% similarUnverified端到端语音架构的早期探索可追溯到2014年前后Google提出的Sequence-to-Sequence模型，以及2016年百度提出的Deep Speech 263% similarUnverifiedAudioPaLM, released by Google in 2023, was the first to inject speech tokens directly into a language model's vocabulary, enabling joint modeling of text and audio in the same representation space.63% similarUnverifiedRT-2, released by Google DeepMind in 2023, was the first to validate the feasibility of fine-tuning robot control on a 55-billion-parameter vision-language model, with generalization capability nearly doubling compared to its predecessor.62% similarUnverifiedGoogle's NotebookLM added personalization settings to chat conversations, offering three modes: default, study guide, and custom.59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/692508API
curl https://kongchang.com/api/v1/knowledge/claims/692508MCP
get_claim(id=692508)