---
language: sv-SE
tags:
- audio
- automatic-speech-recognition
- voxrex
license: cc0-1.0
---

# Wav2vec 2.0 large VoxRex (B)

**Disclaimer:** This is a work in progress.

This model has been pretrained for 200,000 updates on the P4-1k corpus which contains 1000 hours of swedish local public service radio as well as 1500 hours of audio books and other speech from KBs collections.

![Accuracy during training](accuracy.svg "Accuracy")