Automatic Speech Recognition
Transformers
Safetensors
whisper
Inference Endpoints
JackyHoCL's picture
Update README.md
a1a28df verified
metadata
library_name: transformers
license: mit
datasets:
  - AlienKevin/mixed_cantonese_and_english_speech
  - mozilla-foundation/common_voice_17_0
  - mozilla-foundation/common_voice_11_0
metrics:
  - cer
base_model:
  - openai/whisper-large-v3-turbo

CER: 13.7%

transformers-4.46.3

Train Args:
per_device_train_batch_size=16,
gradient_accumulation_steps=1,
learning_rate=1e-5,
gradient_checkpointing=True,
per_device_eval_batch_size=16,
generation_max_length=225,

Hardware:
NVIDIA Tesla V100 16GB * 4

FAQ:

  1. If having tokenizer issue during inference, please update your transformers version to >= 4.46.3
pip install --upgrade transformers==4.46.3