You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

l3-whisper-medium-l3c3_e1_v7

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

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.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: 1
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Wer
0.2112	0.0927	3000	0.4719	22.7174
0.1775	0.1854	6000	0.4478	19.2846
0.16	0.2782	9000	0.4300	17.8644
0.1499	0.3709	12000	0.4118	16.7179
0.1429	0.4636	15000	0.3920	15.8724
0.1354	0.5563	18000	0.3749	15.3029
0.1317	0.6490	21000	0.3774	15.0261
0.1279	0.7417	24000	0.3668	14.1700
0.1247	0.8345	27000	0.3704	13.8788
0.1213	0.9272	30000	0.3650	13.5596

Safetensors

Model size

0.8B params

Tensor type

F32

Base model

Finetuned

(754)

this model