Metadata-Version: 2.5
Name: notepipe-asr-openvino
Version: 0.1.1
Summary: OpenVINO HTTP transcription adapter for notepipe
License-Expression: MIT
Requires-Python: >=3.10
Requires-Dist: huggingface-hub>=0.27
Requires-Dist: numpy>=1.26
Requires-Dist: openvino-genai>=2025.4
Description-Content-Type: text/markdown

# notepipe-asr-openvino

A [notepipe](https://github.com/3ae3ae/notepipe) speech adapter that runs speech recognition locally with
[OpenVINO GenAI](https://github.com/openvinotoolkit/openvino.genai), on an Intel NPU by default.

## Install

```sh
uv tool install "notepipe[openvino]"
```

## Use

```toml
[providers.openvino]
command = "notepipe-asr-openvino"
concurrency = 1

[stages.transcript.speech]
provider = "openvino"
```

| Option | Default | Meaning |
|---|---|---|
| `--model` | `dseditor/Qwen3-ASR-1.7B-INT8_OpenVINO` | Hugging Face repository or local directory of an OpenVINO Qwen3-ASR or Whisper model |
| `--device` | `NPU` | `NPU`, `CPU` or `GPU` |

The first run downloads the model. The transcript stage's `language` is passed to the model.
