| datasets: | |
| - Aniemore/resd | |
| language: | |
| - ru | |
| metrics: | |
| - accuracy | |
| library_name: transformers | |
| pipeline_tag: audio-classification | |
| Finetuned Audio Spectrogram Transformer for sentiment analysis in russian with audio augmentation | |
| [github repo with code and tg bot](https://github.com/glubze-and-tochka/spectrogram-soul) | |
| init state: MIT/ast-finetuned-audioset-10-10-0.4593 |