Unverified50% confidenceBenchmarkExact time
Assuming a correction cost of 5 seconds per transcription error, the correction time per minute of speech after pruning rose from 30 seconds to 64 seconds, a relative increase of +111%.
1
Sources
50%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedResearch shows that the average lag time of top simultaneous interpreters is only 2–3 seconds, while the lag of traditional machine simultaneous interpretation systems often exceeds 5 seconds.74% similarUnverifiedEnd-to-end models have reduced average latency in speech translation from the 2–3 seconds of traditional pipelines to approximately 300 milliseconds.73% similarUnverifiedA streaming pipeline can compress end-to-end latency for voice AI to under 500 milliseconds, whereas the total latency of traditional batch processing can be as high as 3–5 seconds.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949156API
curl https://kongchang.com/api/v1/knowledge/claims/949156MCP
get_claim(id=949156)