This model is a fine-tuned version of facebook/hubert-base-ls960 on the speech_commands dataset. It achieves the following results on the evaluation set:

Loss: 1.1282
Accuracy: 0.8004

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 48
eval_batch_size: 48
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 192
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_ratio: 0.1
num_epochs: 5
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0532	1.0	206	2.3675	0.7068
0.8569	2.0	412	1.4252	0.7864
0.5737	3.0	618	1.3609	0.7959
0.4934	4.0	824	1.2443	0.7999
0.4318	5.0	1030	1.1282	0.8004

Framework versions

Transformers 4.51.2
Pytorch 2.6.0+cu126
Datasets 3.5.0
Tokenizers 0.21.1

gokulsrinivasagan/hubert-base-ls960-speech-commands-v1

hubert-base-ls960-speech-commands-v1

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

gokulsrinivasagan/
hubert-base-ls960-speech-commands-v1