Unverified50% confidenceSolutionExact time
DPSE explicitly decomposes continuous speech features into a semantic pathway and a paralinguistic pathway, which are then fused after dynamic routing and integrated into the language model.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDiffusion language models adopt a parallel denoising approach, which can simultaneously optimize outputs at multiple positions in a single forward pass and then iteratively converge to the final text through multiple rounds.69% similarUnverifiedThe lexical-prosodic unitization and chunking method performs pooling only on consecutive word units to preserve event identity and magnitude in both the text and speech modalities.68% similarUnverifiedRAG 框架将稠密段落检索（Dense Passage Retrieval, DPR）与生成式语言模型结合66% similarUnverifiedExtracting structured fields from unstructured voice text can be achieved through LLM function calling/structured output capabilities, or through traditional Named Entity Recognition (NER) models.64% similarVerifiedNeural audio codecs such as EnCodec and DAC compress continuous audio into discrete token sequences, enabling speech to be directly processed by autoregressive language models just like text tokens.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/982977API
curl https://kongchang.com/api/v1/knowledge/claims/982977MCP
get_claim(id=982977)