From c4df41d209e2377ea06a7cbcead8a92a4a29bcd2 Mon Sep 17 00:00:00 2001 From: Ettore Di Giacinto Date: Tue, 23 Jun 2026 23:44:04 +0000 Subject: [PATCH] chore(recon): bump pins to small-spatial + stem conv kernels (voice 99b1804, face 47fdab6) Measured-gap-driven conv kernels: small-spatial (fill the register tile when output width <= tile width) + small-IC stem + strided-1x1/downsample recovery. ArcFace recognizer 0.57 -> 0.70x MLAS @1t (the closest conv model), WeSpeaker 0.65 -> 0.79x @1t. Parity cosine=1.0 / detect <=1px. The OC-block-sharing lever was a measured dead-end (deep stride-1 is L3-weight-bandwidth bound, not read-port bound) and was NOT shipped. Kernel ceiling reached; further gap needs an algorithm-class change (cache-blocked weight-stationary GEMM, or q8 weights). Signed-off-by: Ettore Di Giacinto Assisted-by: Claude:claude-opus-4-8 [Claude Code] --- backend/go/face-detect/Makefile | 4 ++-- backend/go/voice-detect/Makefile | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/backend/go/face-detect/Makefile b/backend/go/face-detect/Makefile index 677345411..1c1ee4e81 100644 --- a/backend/go/face-detect/Makefile +++ b/backend/go/face-detect/Makefile @@ -1,6 +1,6 @@ # face-detect backend Makefile. # -# Upstream pin lives below as FACEDETECT_VERSION?=d80092b... (.github/bump_deps.sh +# Upstream pin lives below as FACEDETECT_VERSION?=47fdab6... (.github/bump_deps.sh # can find and update it - matches the voice-detect / parakeet.cpp / whisper.cpp # convention). # @@ -14,7 +14,7 @@ # The default target below does the proper clone-at-pin + cmake build so CI does # not need a side-checkout. -FACEDETECT_VERSION?=d80092b1137f739eac814cfe5e1b38b04b003097 +FACEDETECT_VERSION?=47fdab66af84bec8949e683f9f8ab80d7d37253f FACEDETECT_REPO?=https://github.com/mudler/face-detect.cpp GOCMD?=go diff --git a/backend/go/voice-detect/Makefile b/backend/go/voice-detect/Makefile index 278a9e0cc..b70db8f84 100644 --- a/backend/go/voice-detect/Makefile +++ b/backend/go/voice-detect/Makefile @@ -1,6 +1,6 @@ # voice-detect backend Makefile. # -# Upstream pin lives below as VOICEDETECT_VERSION?=7f66871... (.github/bump_deps.sh +# Upstream pin lives below as VOICEDETECT_VERSION?=99b1804... (.github/bump_deps.sh # can find and update it - matches the parakeet.cpp / whisper.cpp / ds4 convention). # # Local dev shortcut: if you already have an out-of-tree voice-detect.cpp build, @@ -13,7 +13,7 @@ # The default target below does the proper clone-at-pin + cmake build so CI does # not need a side-checkout. -VOICEDETECT_VERSION?=7f66871c6dfb9ec6b0ddc329c521c8a9e0e600bb +VOICEDETECT_VERSION?=99b18045295b389c1085c510cc7554b043bb8404 VOICEDETECT_REPO?=https://github.com/mudler/voice-detect.cpp GOCMD?=go