Votic ASR: Fine-tuning Wav2Vec2

Hi all. I don’t speak Votic and I know nothing about it. But it has 7MB of data on CommonVoice so I thought this can be served as a great toy dataset for fast prototyping. What’s more there are hardly any other curated Votic resources on the Internet, so the only way to achieve good performance is leveraging the pertained model effectively. Let’s fine-tune Votic and have fun!