mobedkova
/

wav2vec2-large-xls-r-300m-ru

Automatic Speech Recognition

hf-asr-leaderboard

robust-speech-event

Model card Files Files and versions

Russian Speech Recognition model

Downloads last month: -

Dataset used to train mobedkova/wav2vec2-large-xls-r-300m-ru

Evaluation results

Test WER on Common Voice-7.0
self-reported

27.810
Test CER on Common Voice-7.0
self-reported

8.830
Test WER on Robust Speech Event - Dev Data
self-reported

44.640
Test WER on Robust Speech Event - Test Data
self-reported

42.510

View on Papers With Code