vinoWhisper

Live captions for anything playing on your Linux laptop, running on the NPU that came with it. No cloud, no API key, no account, no audio leaving the machine. Point it at whatever is playing and it captions in your terminal, or in a floating box above everything else on screen. The same box does dictation too: hold a key, talk, and let go to type what you said. There is nothing to configure: start it and it goes.

pip install vinowhisper && vinowhisper-setup && vinowhisper-caption

Why this exists

To be upfront about the bias: I did not build this because local speech-to-text is hard to find. I built it because my laptop has an Intel NPU rated at 16 TOPS that was doing absolutely nothing, and captioning a video turns out to be the rare workload that suits it: continuous, latency-sensitive, and small enough to fit. 16 TOPS is a coprocessor, not a GPU, and whisper-small.en is one of the genuinely useful models that fits in it.

So the pitch is not "another Whisper wrapper." It is: the transcription runs on a chip that is otherwise idle, so it costs you no CPU, no GPU, no fan, and no network. openvino-genai's WhisperPipeline with device="NPU" does the work, the same idea as vinoAuthFace, different feature.

Measured on this laptop (Wildcat Lake, stepping A0), whisper-small.en on the NPU, 2026-09-12:

Per 12s window (the default), generate()	0.70s mean, 0.85s p90
Captions behind the audio	~1.5s at best
First streamed token	0.204s
Idle cost after you stop	zero, the server exits itself

The Intel iGPU does the same window in 0.95s and the CPU in 2.30s, so both fallbacks are usable, just laggier.

What makes it different from a shell script around Whisper

Words are never rewritten once printed. The stitcher runs LocalAgreement-2: a word is committed only when two overlapping windows independently agree on it. That is why the transcript can live in your terminal's own scrollback, where it survives quitting and where your terminal's selection and search still work on it. The hearing… line is the words heard once and still waiting on a second opinion, so the two-cycle commit delay is visible rather than felt as a freeze.

It scales to zero, the systemd way. The NPU model load costs 10-30s, so something has to hold it. A socket unit owns a private Unix socket at boot with no process running, systemd spawns the server on the first connection, and the server exits itself after 5 minutes idle. Serverless, on your laptop, with no framework.

The model download is verified. The Hugging Face source is pinned to a commit and every file is sha256-checked before the export reads it, and the export is then hashed against pins in vinowhisper/model_digests.json before anything loads it. Details in SECURITY.md.

A missing NPU degrades instead of bricking. Selection walks NPU, then GPU, then CPU, and a fallback is never silent: it shows up in the server journal, in /health, in vinowhisper-doctor, and on the status bar as a red border.

It tells you how to fix it. Nearly every error path here prints the command that resolves it, in your distro's package names (see Supported distros).

The overlay is optional, and native. vinowhisper-tray is a 6.3MB Rust binary that floats a caption box above every window, fullscreen video included, with a tray icon, a global shortcut and a start-at-login toggle. It links nothing but libc, adds nothing to the Python install, and draws the same event stream as the terminal UI. It also does dictation: hold Meta+H (the dictation key on laptops that have one), talk, and let go to type what you said into the focused window, one NPU decode per utterance. Optionally, a small model on the NPU tidies it first, and only by deleting: "um", stammers, "three, no wait, four", "scratch that". It never adds or rewords a word, and nothing leaves the machine. docs/tray.md

Install

curl -fsSL https://github.com/karanshukla/vinoWhisper/releases/latest/download/install.sh | bash

That checks the latest release's wheel against its SHA256SUMS, installs it with uv (fetching uv itself, hash-pinned, if you don't have it), and hands over to vinowhisper-setup, which is where every machine-specific decision happens: your capture tool, your NPU driver, the model export your device needs, and systemd units generated against paths that actually exist. It prints every command before running it and asks first.

The desktop overlay (floating captions and dictation) is one more command: vinowhisper-setup --gui. It downloads the binary from the GitHub release and checks it against a digest pinned in the Python package.

To upgrade, use Check for updates on the settings window's About page, or run vinowhisper-upgrade. It installs the new release's wheel, refreshes the overlay and window if you have them, and restarts the server; the exported model stays. To build from a checkout instead, scripts/install.sh still clones and syncs the locked environment.

This is the beginning of the README. Read the rest at the original. Read the full README ↗