Unverified50% confidenceBenchmarkExact time
On a MacBook Air, transcribing one hour of audio takes only about 13 seconds, reaching about 277 times real-time speed.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/13/2026
First Seen
Valid until: 10/27/2026
Sources
Related Entities
Related Claims
UnverifiedFor a cold-start 27k-token fresh session, the original llama.cpp required 245 seconds, the author's fork took 228 seconds on Mac alone, and was reduced to 168 seconds after connecting the phone.72% similarUnverifiedOn a 16G GPU, a 60-second audio clip takes only 101 seconds with 512-frame chunking, 231 seconds with 1024-frame chunking, and 285 seconds without chunking.69% similarUnverifiedMac Computer Use's foreground operations require the user to be idle for at least 2 seconds and wait at most 15 seconds, use a drawing lock, and stop as soon as the user resumes input.67% similarVerifiedThe total chain latency of traditional voice interaction is typically 3–8 seconds, whereas real-time media streaming can compress end-to-end latency to the hundreds-of-milliseconds level.67% similarUnverifiedOn an Apple Silicon MacBook Air (M4, 10-core CPU, 16 GB), Parakeet TDT 0.6B v3 has an average latency of 0.10 seconds, a real-time factor of 0.014, and a WER of 2.50% (25/1002).67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1016769API
curl https://kongchang.com/api/v1/knowledge/claims/1016769MCP
get_claim(id=1016769)