Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
wav2vec2-large-mms-1b-uyghur-latin – AI Model by ixxan | AlphaNeural AI
You can deploy this model and start earning money today!
ixxan
/
wav2vec2-large-mms-1b-uyghur-latin
like
0
transformers
tensorboard
safetensors
wav2vec2
automatic-speech-recognition
generated_from_trainer
ug
facebook/mms-1b-all
finetune
cc-by-nc-4.0
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
wav2vec2-large-mms-1b-uyghur-latin
This model is a fine-tuned version of
facebook/mms-1b-all
on the None dataset. It achieves the following best results on the evaluation set:
Best Wer: 30.8949%
Best Cer: 5.9823 %
Training procedure
Finetuning code avaiblable in
https://github.com/ixxan/ug-speech
Training hyperparameters
The following hyperparameters were used during training:
learning_rate: 0.001
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 32
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 4
mixed_precision_training: Native AMP
Training results
Training Loss
Epoch
Step
Validation Loss
Wer Ortho
Cer Ortho
0.3425
1.0006
1313
0.3081
35.3122
6.8424
0.3218
2.0011
2626
0.2771
31.7204
6.1840
0.3012
3.0017
3939
0.2739
30.8949
5.9823
0.2961
3.9989
5248
0.2771
31.7116
6.1806
Framework versions
Transformers 4.46.2
Pytorch 2.5.1+cu121
Datasets 3.1.0
Tokenizers 0.20.3
