Wav2vec2 Large Slavic Voxpopuli V2

🚀 Wav2Vec2-large-VoxPopuli-V2

This is a large model of Facebook's Wav2Vec2. It is pretrained only in slavic on 88.99999999999999 unlabeled data from the VoxPopuli corpus. The model offers a valuable resource for speech - related tasks in the Slavic language.

🚀 Quick Start

The model is pretrained on 16kHz sampled speech audio. When using the model, ensure that your speech input is also sampled at 16Khz.

⚠️ Important Note

This model does not have a tokenizer as it was pretrained on audio alone. To use this model for speech recognition, a tokenizer should be created and the model should be fine - tuned on labeled text data in slavic. Check out this blog for a more in - detail explanation of how to fine - tune the model.

📚 Documentation

Paper

VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi - Supervised Learning and Interpretation

Authors

Changhan Wang, Morgane Riviere, Ann Lee, Anne Wu, Chaitanya Talnikar, Daniel Haziza, Mary Williamson, Juan Pino, Emmanuel Dupoux from Facebook AI.

More Information

See the official website for more information, here.

📄 License

The license for this model is cc - by - nc - 4.0.

📦 Information Table