mms-akan-ani-kese-v1 – AI Model by teckedd | AlphaNeural AI

You can deploy this model and start earning money today!

teckedd/
mms-akan-ani-kese-v1

like

0

automatic-speech-recognition

generated_from_trainer

facebook/mms-1b-all

endpoints_compatible

Views

No views yet

mms-akan-ani-kese-v1

This model is a fine-tuned version of facebook/mms-1b-all on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.8289
Model Preparation Time: 0.013
Wer: 0.7818
Cer: 0.4587

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: 4
total_train_batch_size: 16
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_steps: 0.1
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Model Preparation Time	Wer	Cer
10.3858	0.4472	250	2.1785	0.013	0.8171	0.4598
9.0871	0.8945	500	1.9688	0.013	0.7977	0.4462
8.2611	1.3417	750	1.8899	0.013	0.7810	0.4456
8.5129	1.7889	1000	1.8479	0.013	0.7851	0.4575
8.2884	2.2361	1250	1.8289	0.013	0.7818	0.4587

Framework versions

Transformers 5.11.0
Pytorch 2.12.0+cu130
Datasets 5.0.0
Tokenizers 0.22.2