| | --- |
| | base_model: facebook/hubert-base-ls960 |
| | library_name: transformers.js |
| | --- |
| | |
| | https://huggingface.co/facebook/hubert-base-ls960 with ONNX weights to be compatible with Transformers.js. |
| |
|
| | ## Usage (Transformers.js) |
| |
|
| | If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@xenova/transformers) using: |
| | ```bash |
| | npm i @xenova/transformers |
| | ``` |
| |
|
| | **Example:** Load and run a `HubertModel` for feature extraction. |
| | ```javascript |
| | import { AutoProcessor, AutoModel, read_audio } from '@xenova/transformers'; |
| | |
| | // Read and preprocess audio |
| | const processor = await AutoProcessor.from_pretrained('Xenova/hubert-base-ls960'); |
| | const audio = await read_audio('https://huggingface.co/datasets/Xenova/transformers.js-docs/resolve/main/jfk.wav', 16000); |
| | const inputs = await processor(audio); |
| | |
| | // Load and run model with inputs |
| | const model = await AutoModel.from_pretrained('Xenova/hubert-base-ls960'); |
| | const output = await model(inputs); |
| | // { |
| | // last_hidden_state: Tensor { |
| | // dims: [ 1, 549, 768 ], |
| | // type: 'float32', |
| | // data: Float32Array(421632) [0.0682469978928566, 0.08104046434164047, -0.4975186586380005, ...], |
| | // size: 421632 |
| | // } |
| | // } |
| | ``` |
| | --- |
| |
|
| | Note: Having a separate repo for ONNX weights is intended to be a temporary solution until WebML gains more traction. If you would like to make your models web-ready, we recommend converting to ONNX using [🤗 Optimum](https://huggingface.co/docs/optimum/index) and structuring your repo like this one (with ONNX weights located in a subfolder named `onnx`). |
