Move standalone repository docs from the root into docs/ and add a documentation index that links the moved docs and the code-local runbooks.
Update Markdown links, source comments, and the root entry points to the new locations. Keep README.md and AGENTS.md at the root as required entry points.
Assisted-by: DeepSeek Harness (deepseek-flash)
At boot whisper-asr could start before the LAN address was configured and
fail with "could not bind on any address". Let whisper-asr.socket bind it
with FreeBind and have uvicorn take fd 3, like the exporters.
Assisted-by: Claude Code (claude-opus-5-5)
The goal is speech recognition on the NPU for vinput, and sherpa-onnx cannot
deliver it: every model it loads keeps a dynamic time axis, which the NPU
compiler rejects outright. Qwen3-ASR and SenseVoice both core-dumped in
vpux-compiler ("Got non broadcastable dimensions pair"), and reshaping the ONNX
inputs does not help because the dynamic dimensions sit inside the graph
(WA-037). Whisper is the only family OpenVINO publishes fully static IRs for,
and its fixed 30 s window is exactly what the compiler accepts.
The host now runs an OpenAI-compatible /v1/audio/transcriptions service backed
by whisper-large-v3-turbo int8 on the NPU: 1.5 s for 11 s of audio against
2.1 s on CPU, with the compiled graph cached in /var/cache so only a cold cache
pays the ~60 s NPU compile. vinput on the desktop sends raw PCM to it through a
small provider script.
Only the NPU driver and compiler still come from the unmerged branch in the
`nixpkgs-npu` input (NixOS/nixpkgs#570230), whose nixpkgs also happens to be
the one where openvino-genai 2026.4.0.0 builds at all.
Assisted-by: DeepSeek Harness 0.2.1-rc.2 (deepseek-v4.1-flash)
Ryan Yin
ryan4yin | 二花