Unverified50% confidenceFactExact time
ASR systems can produce hallucinations, in which the output is fluent and grammatically correct yet entirely unrelated to the actual speech content of the input.
1
Sources
50%
Confidence
Long-term
Relevance
9/8/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedIn traditional voice interaction systems, paralinguistic information carried by speech—such as emotion, tone, and speech rate—is completely lost when converted to plain text during the ASR stage.76% similarUnverifiedThe study proposes that ASR hallucinations are fundamentally a consequence of grounding failure, wherein the transcription is no longer adequately guided by the audio signal.73% similarUnverified现代主流ASR系统多基于Transformer架构，通过在海量音频-文本对上进行监督学习建立声学特征与文字序列的映射关系70% similarUnverifiedIn real-world call recordings with background noise, colloquial expressions, and overlapping speakers, ASR accuracy typically drops below standard benchmarks.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/879152API
curl https://kongchang.com/api/v1/knowledge/claims/879152MCP
get_claim(id=879152)