Automatic Speech Recognition
Transformers
PyTorch
Hindi
wav2vec2
hf-asr-leaderboard
model_for_talk
mozilla-foundation/common_voice_7_0
robust-speech-event
Eval Results (legacy)
# Load model directly
from transformers import AutoProcessor, AutoModelForCTC
processor = AutoProcessor.from_pretrained("Harveenchadha/hindi_base_wav2vec2")
model = AutoModelForCTC.from_pretrained("Harveenchadha/hindi_base_wav2vec2")Quick Links
- Downloads last month
- 14
Model tree for Harveenchadha/hindi_base_wav2vec2
Dataset used to train Harveenchadha/hindi_base_wav2vec2
Evaluation results
- Test WER on Common Voiceself-reported22.620
- Test CER on Common Voiceself-reported7.420
- Test WER on Common Voice-7.0self-reported19.470
- Test CER on Common Voice-7.0self-reported8.050
- Test WER on Common Voice-8.0self-reported20.870
- Test CER on Common Voice-8.0self-reported9.470
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Harveenchadha/hindi_base_wav2vec2")