---
library_name: transformers
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: MSP
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# MSP

This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2884
- Wer: 0.2160
- Cer: 0.1050

## 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: 0.0002
- train_batch_size: 32
- eval_batch_size: 32
- 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: cosine
- lr_scheduler_warmup_steps: 500
- training_steps: 10000

### Training results

| Training Loss | Epoch | Step  | Validation Loss | Wer    | Cer    |
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
| 2.4417        | 0.05  | 500   | 1.5634          | 0.3213 | 0.1614 |
| 2.4653        | 0.1   | 1000  | 1.4770          | 0.2440 | 0.1209 |
| 2.3656        | 0.15  | 1500  | 1.4239          | 0.2407 | 0.1159 |
| 2.4607        | 0.2   | 2000  | 1.6669          | 0.2859 | 0.1395 |
| 2.2601        | 0.25  | 2500  | 1.3392          | 0.2444 | 0.1203 |
| 2.2054        | 0.3   | 3000  | 1.3330          | 0.2428 | 0.1188 |
| 2.0611        | 0.35  | 3500  | 1.8721          | 0.3652 | 0.1965 |
| 2.2652        | 0.4   | 4000  | 1.2884          | 0.2160 | 0.1050 |
| 2.1945        | 0.45  | 4500  | 2.0405          | 0.3451 | 0.1868 |
| 2.4363        | 0.5   | 5000  | 1.4916          | 0.2734 | 0.1337 |
| 2.1200        | 0.55  | 5500  | 1.4868          | 0.2515 | 0.1258 |
| 2.2227        | 0.6   | 6000  | 1.3656          | 0.2379 | 0.1165 |
| 2.0990        | 0.65  | 6500  | 1.4576          | 0.2552 | 0.1291 |
| 2.1397        | 0.7   | 7000  | 1.5793          | 0.2792 | 0.1428 |
| 2.1740        | 0.75  | 7500  | 1.4444          | 0.2380 | 0.1191 |
| 2.3435        | 0.8   | 8000  | 1.4126          | 0.2435 | 0.1231 |
| 2.0578        | 0.85  | 8500  | 1.3806          | 0.2347 | 0.1165 |
| 2.1130        | 0.9   | 9000  | 1.4284          | 0.2449 | 0.1226 |
| 2.1455        | 0.95  | 9500  | 1.4427          | 0.2475 | 0.1233 |
| 2.1259        | 1.0   | 10000 | 1.4531          | 0.2497 | 0.1246 |


### Framework versions

- Transformers 5.10.2
- Pytorch 2.8.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2