Aired in the Ship Log episode of 2026-10-09. This page is the public record for the story: every fact the show was allowed to assert, with its verbatim source sentence — and every reply and correction, unedited.
| Claim | Source sentence (verbatim) | Source |
|---|---|---|
| Cactus released Whistle, a speech recognition model for mobiles, wearables, robots, smart home, automotive and microcontrollers. | “Today we release Whistle, a speech recognition model for mobiles, wearables, robots, smart home, automotive and microcontrollers.” | Hacker News (front page) |
| Whistle is a single 16.9 MB file that runs on the CPU with no dependencies. | “It is one 16.9 MB file, runs on the CPU with no dependencies, and loads into the same C++ engine as Needle, from the same container and the same quantisation.” | Hacker News (front page) |
| Whistle's encoder uses eight Simple Attention blocks with four mHC residual lanes and a Monarch Hadamard MLP replacing the feed-forward network. | “Eight Simple Attention blocks: four mHC residual lanes and a Monarch Hadamard MLP in place of the feed-forward network, the same blocks Needle uses.” | Hacker News (front page) |
| Whistle's decoder has eight Laddered Simple Attention blocks at width 512, with 8 query heads to 2 KV heads. | “Eight Laddered Simple Attention blocks at width 512, 8 query heads to 2 KV heads, 48-dimensional queries and keys, 64-dimensional values, a 3-tap causal convolution on Q, K and V, and engram lookups at layers 3 and 7 over 18,432 slots.” | Hacker News (front page) |
| Whistle's transcript is capped at 320 tokens and its vocabulary is 8,192 text pieces plus seven language tokens. | “The transcript is capped at 320 tokens. The vocabulary is 8,192 text pieces plus seven language tokens, one per language, so the detected language is emitted as a token rather than returned out of band.” | Hacker News (front page) |
| Whistle leads on LibriSpeech test-clean and test-other, SPGISpeech, Earnings-22 and the FLEURS average. | “Whistle is ahead on LibriSpeech test-clean and test-other, on SPGISpeech, on Earnings-22 and on the FLEURS average.” | Hacker News (front page) |
| Whisper base beats Whistle on TED-LIUM, AMI and the MLS average, at 145.3 MB versus 16.9 MB. | “Whisper base is ahead on TED-LIUM, on AMI and on the MLS average, at 145.3 MB against 16.9.” | Hacker News (front page) |
| Whistle's time to first token is 5.9 ms at 5 seconds, 11.1 ms at 10 seconds and 36.3 ms at 30 seconds. | “Whistle's tracks the clip: 5.9 ms at 5 seconds, 11.1 ms at 10, 36.3 ms at 30.” | Hacker News (front page) |
| Whistle's word error rates were measured over 86,174 utterances. | “Whistle's are measured over 86,174 utterances.” | Hacker News (front page) |
| Moonshine is an English-only model. | “Moonshine is English only” | Hacker News (front page) |
Your reply is published verbatim — we do not edit it — and read on air in the next episode, in the same flat voice for every party. We may decline a reply that is abusive, off-topic, or not from the party named; declines are logged. Replies go to a human first; nothing posts automatically.