Instructions to use onnx-community/Voxtral-Mini-3B-2507-ONNX with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers.js
How to use onnx-community/Voxtral-Mini-3B-2507-ONNX with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('audio-text-to-text', 'onnx-community/Voxtral-Mini-3B-2507-ONNX');
Add Timestamp Support
#5
by flackzz - opened
Is it possible to implement timestamp support here?
According to the official Mistral documentation on audio transcription, Voxtral supports timestamps through the Mistral API.
However, this feature is currently not available in the Hugging Face Transformers implementation of the Voxtral models.
Timestamps are a crucial feature for many ASR use cases and without timestamp support, Voxtral’s potential in real-world scenarios is limited.