Unverified50% confidenceFactExact time
ASR技术最初基于隐马尔可夫模型（HMM），通过维特比算法在状态路径中寻找后验概率最大的路径将声学特征映射为音素串
1
Sources
50%
Confidence
Long-term
Relevance
7/20/2026
First Seen
Sources
Related Claims
UnverifiedHMM假设语音波形由音素状态序列依概率生成，解码时通过维特比算法寻找后验概率最大的状态路径，并通常与高斯混合模型结合形成GMM-HMM架构82% similarUnverifiedAutomatic Speech Recognition (ASR) at the underlying level typically employs end-to-end neural network models such as Whisper and Wav2Vec.72% similarVerifiedSpeech recognition technology has evolved from hidden Markov models (HMMs) to recurrent neural networks (RNNs) and then to end-to-end models.70% similarUnverifiedEarly mainstream speech recognition solutions relied on Hidden Markov Models (HMM), which had limited capabilities for noise, accent, and multilingual mixed recognition.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/573326API
curl https://kongchang.com/api/v1/knowledge/claims/573326MCP
get_claim(id=573326)