---
library_name: transformers
license: apache-2.0
base_model: facebook/wav2vec2-large-xlsr-53
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: xlsr-a-nose
  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. -->

# xlsr-a-nose

This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3950
- Wer: 0.3216

## 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: 0.0004
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Use 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: 132
- num_epochs: 100
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Wer    |
|:-------------:|:-------:|:----:|:---------------:|:------:|
| 4.473         | 2.7027  | 200  | 2.4188          | 1.0    |
| 1.3356        | 5.4054  | 400  | 0.2886          | 0.5197 |
| 0.1766        | 8.1081  | 600  | 0.3087          | 0.3930 |
| 0.0958        | 10.8108 | 800  | 0.2624          | 0.3323 |
| 0.0596        | 13.5135 | 1000 | 0.3792          | 0.3440 |
| 0.0461        | 16.2162 | 1200 | 0.3037          | 0.3365 |
| 0.0352        | 18.9189 | 1400 | 0.3387          | 0.3387 |
| 0.0227        | 21.6216 | 1600 | 0.3182          | 0.3387 |
| 0.0243        | 24.3243 | 1800 | 0.3493          | 0.3450 |
| 0.0224        | 27.0270 | 2000 | 0.3503          | 0.3312 |
| 0.0145        | 29.7297 | 2200 | 0.3551          | 0.3301 |
| 0.0205        | 32.4324 | 2400 | 0.3310          | 0.3514 |
| 0.0126        | 35.1351 | 2600 | 0.3741          | 0.3323 |
| 0.0137        | 37.8378 | 2800 | 0.3761          | 0.3323 |
| 0.014         | 40.5405 | 3000 | 0.3646          | 0.3355 |
| 0.0096        | 43.2432 | 3200 | 0.3660          | 0.3301 |
| 0.0099        | 45.9459 | 3400 | 0.3745          | 0.3312 |
| 0.0121        | 48.6486 | 3600 | 0.3930          | 0.3291 |
| 0.0131        | 51.3514 | 3800 | 0.3886          | 0.3280 |
| 0.0095        | 54.0541 | 4000 | 0.3874          | 0.3450 |
| 0.0066        | 56.7568 | 4200 | 0.3877          | 0.3269 |
| 0.0047        | 59.4595 | 4400 | 0.3871          | 0.3248 |
| 0.0074        | 62.1622 | 4600 | 0.3886          | 0.3259 |
| 0.0068        | 64.8649 | 4800 | 0.4206          | 0.3259 |
| 0.005         | 67.5676 | 5000 | 0.4182          | 0.3227 |
| 0.0087        | 70.2703 | 5200 | 0.4100          | 0.3248 |
| 0.0047        | 72.9730 | 5400 | 0.4196          | 0.3259 |
| 0.0056        | 75.6757 | 5600 | 0.4133          | 0.3259 |
| 0.0048        | 78.3784 | 5800 | 0.4135          | 0.3269 |
| 0.0036        | 81.0811 | 6000 | 0.3901          | 0.3248 |
| 0.003         | 83.7838 | 6200 | 0.3869          | 0.3227 |
| 0.0021        | 86.4865 | 6400 | 0.3896          | 0.3227 |
| 0.0013        | 89.1892 | 6600 | 0.3893          | 0.3216 |
| 0.0019        | 91.8919 | 6800 | 0.3983          | 0.3216 |
| 0.0015        | 94.5946 | 7000 | 0.4023          | 0.3227 |
| 0.0013        | 97.2973 | 7200 | 0.3973          | 0.3227 |
| 0.0017        | 100.0   | 7400 | 0.3950          | 0.3216 |


### Framework versions

- Transformers 4.47.0.dev0
- Pytorch 2.4.0
- Datasets 3.0.1
- Tokenizers 0.20.0