---
library_name: transformers.js
pipeline_tag: automatic-speech-recognition
license: apache-2.0
base_model: primeline/distil-whisper-large-v3-german
tags:
- transformers.js
- onnx
- whisper
- speech
- automatic-speech-recognition
- timestamps
---

# distil-whisper-large-v3-german_timestamped-ONNX

This repository contains ONNX weights for [`primeline/distil-whisper-large-v3-german`](https://huggingface.co/primeline/distil-whisper-large-v3-german)
prepared for use with Transformers.js.

Timestamp support is preserved through the exported Whisper generation config.
Available dtypes in this export: `fp32, fp16, q4, q8`.

The repository layout follows the standard Transformers.js Whisper convention:

- processor, tokenizer, and config files in the repository root
- encoder and merged decoder ONNX files inside `onnx/`

## Usage (Transformers.js)

```js
import { pipeline } from "@huggingface/transformers";

const transcriber = await pipeline("automatic-speech-recognition", "<your-hf-username>/distil-whisper-large-v3-german_timestamped-ONNX", {
  device: "webgpu",
  dtype: "fp16", // or "fp32", "q4", "q8"
});

const output = await transcriber("audio.mp3", {
  chunk_length_s: 30,
  return_timestamps: "word",
});

console.log(output.text);
console.log(output.chunks?.slice(0, 3));
```

Recommended choices:

- `fp32`: highest precision, typically for server-side or high-memory WebGPU
- `fp16`: smaller WebGPU model with good speed/quality tradeoff
- `q8`: smaller CPU/WASM model
- `q4`: smallest model, best for constrained devices

## Source Model

- Base model: [`primeline/distil-whisper-large-v3-german`](https://huggingface.co/primeline/distil-whisper-large-v3-german)
- Export format: ONNX
- Intended runtime: [Transformers.js](https://huggingface.co/docs/transformers.js)