tiny-audio-multitask

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.8849
• Train Loss Sift: 1.8841
• Train Loss Transcription: 1.8856

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: 0.001
• train_batch_size: 14
• eval_batch_size: 14
• seed: 42
• 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: polynomial
• lr_scheduler_warmup_steps: 1000
• num_epochs: 4
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Loss Sift	Loss Transcription
2.0532	0.5636	5000	2.0112	2.0122	2.0102
2.0046	1.1273	10000	1.9566	1.9575	1.9557
1.9870	1.6909	15000	1.9285	1.9228	1.9342
1.9600	2.2545	20000	1.9108	1.9076	1.9142
1.9574	2.8182	25000	1.8970	1.8988	1.8953
1.9567	3.3818	30000	1.8913	1.8873	1.8950
1.9488	3.9454	35000	1.8849	1.8841	1.8856

Framework versions

• Transformers 5.0.0.dev0
• Pytorch 2.8.0+cu128
• Datasets 3.6.0
• Tokenizers 0.22.2