metadata
language:
- uk
license: cc-by-nc-sa-4.0
datasets:
- mozilla-foundation/common_voice_10_0
Join Ukrainian Speech Recognition Community - https://t.me/speech_recognition_uk
This model was trained using the base model https://huggingface.co/fav-kky/wav2vec2-base-cs-80k-ClTRUS (pre-trained from 80 thousand hours of Czech speech)
This model has apostrophes and hyphens.
Metrics (CV7 test split):
Without LM:
- WER: 0.4416
- CER: 0.1042
With LM:
- WER: 0.1422
- CER: 0.0464
Metrics (CV10 test split):
Without LM:
- WER: 0.4121
- CER: 0.0946
With LM:
- WER: 0.0981
- CER: 0.0328