granite-speech-2b-za-multilingual

This model is a fine-tuned version of ibm-granite/granite-speech-3.3-2b on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.1306

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: 3e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.2
• training_steps: 10000

Training results

Training Loss	Epoch	Step	Validation Loss
21.9822	0.1	1000	2.6979
20.0781	0.2	2000	2.3706
20.5725	0.3	3000	2.1988
18.4666	0.4	4000	2.1487
18.3545	0.5	5000	2.1397
17.3601	1.1	6000	2.1420
18.1428	1.2	7000	2.1543
19.9344	1.3	8000	2.1296
18.2329	1.4	9000	2.1209
18.2982	1.5	10000	2.1306

Framework versions

• Transformers 4.57.3
• Pytorch 2.9.1+cu128
• Datasets 4.4.1
• Tokenizers 0.22.1