Automatic Speech Recognition
Transformers
Safetensors
Ukrainian
wav2vec2-bert
audio
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForCTC
processor = AutoProcessor.from_pretrained("speech-uk/w2v-bert")
model = AutoModelForCTC.from_pretrained("speech-uk/w2v-bert")Quick Links
w2v-bert-uk v2.1
Overview
This is the model - https://huggingface.co/Yehor/w2v-bert-uk-v2.1 - where tensors are saved in BF16 format.
Community
- Discord: https://bit.ly/discord-uds
- Speech Recognition: https://t.me/speech_recognition_uk
- Speech Synthesis: https://t.me/speech_synthesis_uk
See other Ukrainian models: https://github.com/egorsmkv/speech-recognition-uk
- Downloads last month
- 15
Model tree for speech-uk/w2v-bert
Base model
facebook/w2v-bert-2.0Dataset used to train speech-uk/w2v-bert
Spaces using speech-uk/w2v-bert 2
Evaluation results
- WER on common_voice_10_0test set self-reported17.340
- CER on common_voice_10_0test set self-reported3.330
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="speech-uk/w2v-bert")