Downloads · 30 days
125
16% of all-time downloads
whisper_large_CGN
kul-speech-lab/whisper_large_CGN
whisper_large_CGN is a automatic speech recognition model from kul-speech-lab. Use it when you need speech turned into text. It is set up for transformers. The card lists the license as apache-2.0.
This model is a fine-tuned version of openai/whisper-large-v2 on the kul-speech-lab/CGN dataset. It achieves the following results on the evaluation set: - Loss: 0.23932012915611267 - Wer: 9.615871912312803
All-time downloads
806
Public
Repo size
18.5 GB
Likes
3
Trending 1
Repo makeup
Click a slice to open those files.
.pt12.3 GB · 67%
At a glance
- Task
- Automatic Speech Recognition
- Library
- transformers
- License
- apache-2.0
- Model type
- whisper
- Access
- Public
- Created
- Apr 25, 2023
- Updated
- Apr 25, 2023
- SHA
- b9e18f60
Datasets
Model card
From the Hugging Face model README
Whisper Large CGN
This model is a fine-tuned version of openai/whisper-large-v2 on the kul-speech-lab/CGN dataset. It achieves the following results on the evaluation set:
- Loss: 0.23932012915611267
- Wer: 9.615871912312803
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 32
- eval_batch_size: 16
- gradient_accumulation_steps: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 15000
- mixed_precision_training: Native AMP
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2
Whisper large model finetuned on Flemish part of Corpus Gesproken Nederlands (CGN).
- Task
- Automatic Speech Recognition
- Library
- transformers
- Type
- whisper
- License
- apache-2.0
- Created
- Apr 25, 2023
- Updated
- Apr 25, 2023