Instructions to use AlanRobotics/whisper-tiny-ru with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AlanRobotics/whisper-tiny-ru with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="AlanRobotics/whisper-tiny-ru")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("AlanRobotics/whisper-tiny-ru") model = AutoModelForSpeechSeq2Seq.from_pretrained("AlanRobotics/whisper-tiny-ru", device_map="auto") - Notebooks
- Google Colab
- Kaggle
metadata
license: apache-2.0
datasets:
- mozilla-foundation/common_voice_11_0
language:
- ru
metrics:
- wer
library_name: transformers
pipeline_tag: automatic-speech-recognition
Model
Модель whisper- tiny (39 M параметров), зафайнтюненная для русского языка
Fine- tuning
learning epochs: 1
learning rate: 1e-6
parameters: 39 M
Evaluation
Тестовый датасет common_voice_11_0, ru
AlanRobotics/whisper-tiny-ru WER: 39.9
openai/whisper-tiny WER: 56.5