---
license: apache-2.0
---

🇺🇦 Join Ukrainian Speech Recognition Community - https://t.me/speech_recognition_uk

⭐ See other Ukrainian models - https://github.com/egorsmkv/speech-recognition-uk

This is a pre-trained Ukrainian wav2vec2 XLS-R model with 300m parameters (dataset is 323h, source of speech is **broadcast** programs).  

Steps: 400,000

The model is not intended to do inference, it's only for fine-tuning on own labeled dataset.

The model was trained from [the wav2vec2 XLS-R model](https://huggingface.co/facebook/wav2vec2-xls-r-300m) with 300m parameters.